JPWO2002035665A1 - Optical transmitter, optical repeater, optical receiver, and optical transmission method - Google Patents

Abstract

An optical signal generating means (20) for generating a main signal to be transmitted and an inverted signal thereof as optical signals of a plurality of wavelengths; and a plurality of optical signals generated by the optical signal generating means. Wavelength multiplexing means (21D) for multiplexing and transmitting optical signals of different wavelengths, and transmitting a main signal to be transmitted and its inverted signal as wavelength multiplexed optical signals of a plurality of wavelengths, A crosstalk between channels generated when multi-wavelength collective amplification is performed by an optical amplifier such as a Raman amplifier or a semiconductor optical amplifier can be effectively suppressed by a method independent of the performance and characteristics of an optical device.

Description

）で変調されるようになっている。なお、これらの外部変調器２１Ｂ−ｉには、後述するように例えば公知のマッハツェンダ型光変調器が適用される。
また、各可変減衰器２１Ｃ−ｉは、それぞれの減衰度が調整されることにより、対応する変調器２１Ｂ−ｉの出力レベルを個々に調整して光合波器２１Ｄへの光信号の入力レベルを調整するためのもので、具体的には、光合波器２１Ｄへの各光信号の入力レベルが一様になるようそれぞれの減衰度が調整される。さらに、光合波器（波長多重手段）２１Ｄは、これらの各可変減衰器２１Ｃ−ｉの出力を合波（ｎ波長多重）してＷＤＭ信号としてＥＤＦＡ２２へ出力（送信）するものである。
このような構成により、本光合波部２１では、まず、外部変調器２１Ｂ−（２ｋ−１）にて、光源２１Ａ−（２ｋ−１）からの光信号（波長λ_２ｋ−１）が信号Ｑｋで変調され、外部変調器２１Ｂ−２ｋにて、光源２１Ａ−２ｋからの光信号（波長λ_２ｋ）が反転信号Ｑｋバーで変調される。
この結果、例えば図４（Ａ）及び図４（Ｂ）に模式的に示すように、外部変調器２１Ｂ−（２ｋ−１）からは、信号Ｑｋの情報をもつ光信号（波長λ_２ｋ−１）が出力され、外部変調器２１Ｂ−２ｋからは、その信号Ｑｋのマーク（ビット値“１”）とスペース（ビット値“０”）とを反転した信号Ｑｋバーの情報をもつ光信号（波長λ_２ｋ）が出力される。
そして、これらの各光信号ＱｋとＱｋバーは、それぞれ、対応する可変減衰器２１Ｃ−ｉにてその光レベルが全波長λ１〜λｎで一様になるよう調整されたのち、光合波器２１Ｄにて合波（ｎ波長多重）されてＷＤＭ信号としてＥＤＦＡ２２及び光合波器２４を通じて光ファイバ伝送路５−１へ出力される。
つまり、本実施形態の光合波部２１は、信号Ｑ_１とその反転信号Ｑ_１バーは波長λ１とλ２、信号Ｑ_２とその反転信号Ｑ_２バーは波長λ３とλ４といった具合に、同じ情報内容の信号Ｑｋ（Ｑｋバー）を送信するのに異なる２つの波長λ_２ｋ−１，λ_２ｋを使って送信するようになっているのである。即ち、上記の光源２１Ａ−ｉ，外部変調器２１Ｂ−ｉ及び可変減衰器２１Ｃ−ｉは、図２中に示すように、信号Ｑｋとその反転信号Ｑｋバーとを２種類の波長λ_２ｋ−１，λ_２ｋの光信号として生成するための光信号生成手段２０を構成していることになる。
これにより、これら２つの波長λ_２ｋ−１，λ_２ｋの光信号は、図４（Ａ）及び図４（Ｂ）に模式的に示すように、光ファイバ伝送路５−１，５−２中を「前方向励起」用のラマン励起光Ｐ（波長λ０）とともに伝播することになるが、このとき、２つの波長λ_２ｋ−１，λ_２ｋの光信号がほぼ同期関係を保ったまま伝播するとすれば、図４（Ｃ）に模式的に示すように、それぞれの波長λ_２ｋ−１，λ_２ｋの光信号パワーの合計が常にほぼ一定となる。
この結果、図４（Ｃ）中に示すように、ラマン増幅時に信号Ｑｋ及びＱｋバーに消費されるラマン励起光Ｐのエネルギー量が一様になるので、ラマン励起光Ｐに対する変調効果が抑制されて、「前方向励起」において顕著に現われる「チャンネル間クロストーク」を効果的に抑圧することが可能となる。
ここで、ラマン増幅媒体となる光ファイバ伝送路５−１，５−２の波長分散による信号Ｑｋとその反転信号Ｑｋバーとの位相ずれ（遅延差）について見積もってみる。光ファイバ伝送路５−１，５−２に、分散シフトファイバや非ゼロ分散シフトファイバを使った場合を想定すると、光ファイバ伝送路５−１，５−２は、１ｐｓ／ｋｍ／ｎｍ程度の分散量をもつことになる。
従って、例えば、隣接チャンネル（波長）間隔を１ｎｍ、伝送距離を１００ｋｍ（キロメートル）と仮定すると、上記の分散による信号Ｑｋとその反転信号Ｑｋバーとのずれは１００ｐｓとなる。この遅延差は、１０Ｇｂｐｓの信号速度の１タイムスロット分、伝送路長にして約３ｃｍ（センチメートル）に相当する。
しかしながら、「前方向励起」によるラマン増幅効果が起こるのは、図１８（Ｄ）に模式的に示すように、送信端に近い部分が支配的であるため、少なくとも、送信端の部分では十分に同期関係が保たれ、有効にクロストークが抑圧されると考えられる。
ただし、上記の信号Ｑｋとその反転信号Ｑｋバーとを合波するまでに上記１タイムスロット分、つまり、３ｃｍ以上の光路差があると送信端で既に同期関係が全くとれない状態になってしまうことになる。このため、少なくとも、本光合波部２では、上記の２つの信号Ｑｋ及びＱｋバーの組がそれぞれ互いに位相同期した状態で光合波器２１Ｄにて波長多重されるようにする必要がある。
そこで、本実施形態では、図２中に○印や△印を付して示すように、少なくとも、外部変調器２１Ｂ−（２ｋ−１）から光合波器２１Ｄまでの光路長Ｌ_２ｋ−１と、外部変調器２１Ｂ−２ｋから光合波器２１Ｄまでの光路長Ｌ_２ｋとが同じ光路長、即ち、外部変調器２１Ｂ−（２ｋ−１）及び外部変調器２１Ｂ−２ｋの組毎に、光合波器２１Ｄまでの光路長がそれぞれ同じとなるように、各外部変調器２１Ｂ−ｉから光合波器２１Ｄまでの光路長（配置）を設計する。なお、勿論、全ての外部変調器２１Ｂ−ｉと光合波器２１Ｄとの間の光路長を同じにしてもよい。
これにより、信号ＱｋとＱｋバーとの組を互いに位相同期した状態で光合波器２１Ｄにて合波して送信することができ、本光合波部２１（送信局２）の特性改善を図って、「チャンネル間クロストーク」の抑圧効果を最大限に発揮させることができる。
以上のように、本実施形態の送信局２によれば、送信すべき信号Ｑｋの波形を反転した信号Ｑｋバーを生成し、これらの信号Ｑｋと信号Ｑｋバーとを隣接する２つの波長λ_２ｋ−１，λ_２ｋを使って同期関係を保ったまま送信することにより、「前方向励起」によるラマン増幅時に顕著に発生する「チャンネル間クロストーク」を、光デバイスの性能や特性に依存することなく効果的に抑圧することができ、結果として、従来の２倍以上の長距離伝送が可能になる。
従って、同じ伝送距離であれば、システム１に必要とされる中継局数を従来よりも大幅に削減することができて、システム１のコストダウンを図ることができるし、同じ中継局数であれば、従来の２倍以上の長距離伝送が可能なシステム１を構築できることになる。
また、図３により上述したように、本実施形態では、送信すべき信号Ｑｋを反転ゲート２１Ｅにて電気信号のまま反転して反転信号Ｑｋバーを得、この反転信号Ｑｋバーで波長λ_２ｋの光信号を変調することによって反転した光信号を得るので、反転した光信号を得るにあたって、既存の光送信機の基本構成や光部品に関しては何ら変更を加えることなく、電気回路のみの改良で済むので、容易に適用（実現）可能である。
さらに、外部変調器２１−ｉと光合波器２１Ｄとの間には、それぞれ、可変減衰器２１Ｃ−ｉが設けられているので、信号Ｑｋとその反転信号Ｑｋバーの各光レベル（パワー）を個々に調整することができ、これらの信号Ｑｋとその反転信号Ｑｋバーの合計パワーを「チャンネル間クロストーク」の抑圧効果が最大限に発揮される最適な状態に制御することも可能である。
なお、本実施形態では、同じ情報内容の信号Ｑｋ（Ｑｋバー）に対して２つの波長λ_２ｋ−１，λ_２ｋを用いるため、必然的に、従来のように１つの信号に対して１波長を割り当てる場合に比して、波長帯域を半分しか使えないことになり、有益度が低くなるのではないのかという指摘があるかもしれない。以下、この点について考察する。
ＷＤＭ光伝送方式において一般に多重度を上げる方法としては、光増幅器の増幅帯域を広げることと波長間隔を小さくすることが挙げられる。波長間隔は、現状装置では、例えば１００ＧＨｚ（ギガヘルツ）間隔が主に採用されており、次世代新機種装置では、その半分の５０ＧＨｚ間隔やさらにその半分の２５ＧＨｚ間隔に設定するなど、多重度をさらに上げてゆく方向にあると考えられる。
そこで、上述のごとく同じ情報内容の信号Ｑｋ及びＱｋバーの送信に２波長λ_２ｋ−１，λ_２ｋを使う本方式を採用して、「チャンネル間クロストーク」を抑圧して長距離伝送を実現しつつ、多重度を落とさないためには、この波長間隔をさらに狭める方法が考えられる。では、どこまで、波長間隔を狭めることができるのかという点について考えてみる。
波長間隔を狭くしていった場合、伝送特性を制限する要因は線形クロストークと非線形クロストークとに分けられる。このうち、線形クロストークについては合分波器の隣接チャンネルパワーの漏れ込み等によるものが考えられ、これについては本方式を採用する如何に関わらず発生する。
これに対し、非線形クロストークは、前述したラマン増幅によるものの他にも、自己位相変調効果（ＳＰＭ）、相互位相変調効果（ＸＰＭ），４光波混合（ＦＷＭ）によるものがある。ここで、信号Ｑｋとその反転信号Ｑｋバーとの間の波長間隔を限りなく狭くして、ほぼ１つの波長にこれらの２つの信号Ｑｋ，Ｑｋバーが載っているとみなしてそれぞれのパワーを合計すると限りなく直流に近いパワーが流れていることになる。このため、この１波長分とみなした信号から他チャンネルに影響するクロストークは直流に近いものとなるはずである。
従って、本方式を採用することにより、ＳＰＭ，ＸＰＭ，ＦＷＭなどの非線形クロストークも抑圧される方向に働くものと期待される。さらに、波長間隔が限りなく小さくなった場合には、波長分散による信号Ｑｋ，Ｑｋバー間の位相ずれもそれに比例して小さくなるため、クロストーク抑圧効果はさらに大きくなると期待される。
よって、複数波長にそれぞれ異なる信号を載せて送信する従来方式で波長多重度を高めるよりも、本方式のように、波長多重される全波長のうち信号Ｑｋとその反転信号Ｑｋバーとが隣接波長で交互に含まれる信号を多重化した方が、より低雑音で長距離伝送可能な光伝送システムを容易に実現できると期待される。
（Ｂ）反転信号生成方法の第１変形例の説明
なお、図３に示す回路（光信号生成手段２０）は、例えば図５に示す構成としてもよい。即ち、外部変調器２１Ｂ−（２ｋ−１）と外部変調器２１Ｂ−２ｋの組毎に、バイアス制御回路２１３を設けるとともに、反転ゲート２１Ｅは用いずに、外部変調器２１Ｂ−（２ｋ−１），２１Ｂ−２ｋのそれぞれには同じ電気信号の信号Ｑｋを入力する構成とする。
そして、バイアス制御回路２１３から、例えば図６に模式的に示すように、外部変調器２１Ｂ−（２ｋ−１），２１Ｂ−２ｋ〔信号Ｑｋの光導波路に設けられた電極（図示省略）〕に供給（印可）するバイアス電圧を制御することで、上記光導波路の光透過率を制御して、一方の外部変調器２１Ｂ−（２ｋ−１）の出力ポートから信号Ｑｋ（実線５２参照）が出力されるとともに、他方の外部変調器２１Ｂ−２ｋの出力ポートから信号Ｑｋが反転して出力（破線５３参照）されるような状態で外部変調器２１Ｂ−（２ｋ−１），２１Ｂ−２ｋを駆動する。
なお、図６において、実線５０は外部変調器２１Ｂ−（２ｋ−１）に印可されるバイアス電圧、破線５１は外部変調器２１Ｂ−２ｋに印可されるバイアス電圧をそれぞれ表わす。
つまり、上記のバイアス制御回路２１３は、一方の外部変調器２１Ｂ−（２ｋ−１）から光信号としての信号Ｑｋが出力されるとともに他方の外部変調器２１Ｂ−２ｋから光信号としての反転信号Ｑｋバーが出力されるように各外部変調器２１Ｂ−（２ｋ−１），２１Ｂ−２ｋの変調状態を制御する変調状態制御回路として機能するのである。
これにより、この場合は、反転信号Ｑｋバーを得るのに、上述したように電気信号で反転する必要が無い（反転ゲート２１Ｅが必要無い）ため、低コストおよび小型化が可能である。また、電気信号のパスが異なる（反転ゲート２１Ｅを通るか通らないかなど）ことに起因する信号Ｑｋとその反転信号Ｑｋバーとの間の遅延の発生を回避して、信号Ｑｋと反転信号Ｑｋバーとをより同期した状態で送信することができる。
（Ｃ）反転信号生成方法の第２変形例の説明
また、図３に示す回路（光信号生成手段２０）は、例えば図７に示す構成としてもよい。即ち、外部変調器２１Ｂ−（２ｋ−１），２１Ｂ−２ｋとして、２つのマッハツェンダ型光変調器を並列に並べて、同じ側の入力ポートに光源２１Ａ−（２ｋ−１），２１Ａ−２ｋからのそれぞれ異なる波長λ_２ｋ−１，λ_２ｋの光信号を入力し、それぞれの電極２１１，２１２に送信すべき信号（電気信号）Ｑを変調信号として供給する。なお、この図７には、代表例として、外部変調器２１Ｂ−１及び２１Ｂ−２（波長λ１および波長λ２）に着目した構成を示している。
これにより、出力ポートとしてそれぞれ反対（図７では外部変調器２１Ｂ−１の出力ポート「２」と外部変調器２１Ｂ−２の出力ポート「１」）のものを使えば、信号Ｑ_１とその反転信号Ｑ_１バーとが得られる。なお、マッハツェンダ型光変調器自体の動作については公知である。
また、図７中に破線で示すように、これらの２つの信号Ｑｋ及びＱｋバー（外部変調器２１Ｂ−１の出力ポート「２」と外部変調器２１Ｂ−２の出力ポート「１」）を光合波器２１３で合波する構成にすれば、外部変調器２１Ｂ−１，２１Ｂ−２と光合波器２１３とを一体化（マッハツェンダ型光変調・合波器として構成）して、１つの基板上に集積化することも可能である。
このように、外部変調器２１Ｂ−（２ｋ−１）及び２１Ｂ−２ｋに、マッハツェンダ型光変調器を用いることにより、送信局２に必要とされる変調器２１−ｉを非常に簡単且つ小型で実現することができ、光合波部２１ひいては送信局２の大幅な小型化が可能である。また、光合波器２１３を用いることにより、信号Ｑｋと反転信号Ｑｋバーとの間の遅延差を最小にすることが可能であり、「チャンネル間クロストーク」の抑圧効果をより効果的に発揮させることができる。
（Ｄ）反転信号生成方法の第３変形例の説明
次に、上記の光信号生成手段２０において反転信号Ｑｋバーを得るには、例えば図８に示すように、半導体光増幅器２１Ｆ−ｋ（ｋ＝１〜ｎ）を利用する方法も考えられる。即ち、半導体光増幅器２１Ｆ−ｋに、送信すべき信号で変調した光源２１Ａ−（２ｋ−１）からの波長λ_２ｋ−１の信号と、もう１つの光源２１Ａ−２ｋからの波長λ_２ｋの直流信号とを光合波器２１５で合波した上で入力する。
つまり、この場合の光源２１Ａ−（２ｋ−１）は、光信号としての信号Ｑｋを（直接変調方式により）生成する主信号生成回路として機能し、光源２１Ａ−２ｋは、光信号としての直流信号を生成する直流信号生成回路として機能する。
そして、半導体光増幅器２１Ｆ−ｋの利得を利得制御回路２１４によって制御して、半導体光増幅器２１Ｆ−ｋを利得飽和状態で動作させる。すると、前述した半導体光増幅器２１Ｆ−ｋのもつクロストーク特性により、波長λ_２ｋの直流信号が変調される。このとき、この変調波が信号Ｑｋの反転波となるように半導体光増幅器２１Ｆ−ｋの利得を調整することで、反転信号Ｑｋバーを得ることができる。
つまり、半導体光増幅器２１Ｆ−ｋにより、上記の直流信号の強度が信号Ｑｋの波形に応じて変調されることを利用して（半導体光増幅器２１Ｆ−ｋを変調器として利用して）、光信号としての信号Ｑｋとその反転信号Ｑｋバーとを得るのである。
従って、この場合も、電気信号で信号を反転する必要が無く、変調器として機能する半導体光増幅器も隣接波長λ_２ｋ−１，λ_２ｋの組毎にそれぞれ１台ずつで済むため、低コスト化および小型化が可能である。また、信号Ｑｋとその反転信号Ｑｋバーとの間の遅延も発生しない。
さらに、この場合は、入力光を電気信号に変換することなくそのまま入力として扱うことができるため、光源２１Ａ−ｉを使用せずに構成することも可能である。従って、例えば、光クロスコネクト装置や光ＡＤＭ（Ａｄｄ−Ｄｒｏｐ　Ｍｕｌｔｉｐｌｅｘｅｒ）などにおいて扱われる光信号をそのまま入力として扱うこともできる。
（Ｅ）光合波部２１の第１変形例の説明
次に、ここでは、図２に示した光合波部２１の第１変形例について説明する。
図２により上述した光合波部２１では、波長数ｎ分の可変減衰器２１Ｃ−ｉを設けているが、通常、光ファイバ伝送路５−１，５−２は波長依存性の伝送損失特性を有しているため、信号Ｑｋとその反転信号Ｑｋバーとを隣接する波長λ_２ｋ−１，λ_２ｋを用いて送信する場合には、その波長λ_２ｋ−１，λ_２ｋの違いによる光ファイバ伝送路５−１，５−２における伝送損失の違いは軽微であると考えられる。
つまり、隣接する波長λ_２ｋ−１，λ_２ｋであれば制御すべき伝送損失値も大きく異なるは無く、これらの隣接波長λ_２ｋ−１，λ_２ｋの光信号を一括して制御することによる特性の劣化は小さいものと考えられるため、光合波部２１では、必ずしも、隣接波長λ_２ｋ−１，λ_２ｋ毎に減衰度（光送信パワー）制御を行なう必要は無い。
そこで、信号Ｑｋとその反転信号Ｑｋバーとを隣接波長λ_２ｋ−１，λ_２ｋを使って送信する場合は、例えば図９に示すように、光信号生成手段２０において、変調器２１Ｂ−（２ｋ−１）と変調器２１Ｂ−２ｋとの組毎に光結合器２１Ｇ−ｋを設けて、変調器２１Ｂ−（２ｋ−１）の出力と変調器２１Ｂ−２ｋの出力とを出力直後に光結合器２１Ｇ−ｋにて結合する構成を採る。そして、可変減衰器２１Ｃ−ｋにおいては、２波長λ_２ｋ−１，λ_２ｋ分の光信号レベルを一括制御すればよい。
ただし、この場合（あるいは、図７や図８により上述した構成を適用した場合）は、１つの可変減衰器２１Ｃ−ｉの出力に、複数（２つ）の波長（チャンネル）λ_２ｋ−１，λ_２ｋ−１が含まれることになるので、例えば図１３に模式的に示すように、１チャンネル当たりの通過帯域に２波長λ_２ｋ−１，λ_２ｋ−１の光信号が含まれるように設計された、１チャンネル当たりの通過帯域が通常のものよりも広帯域な光合波器２１Ｄ′を用いる。これにより、複数の波長を含む光信号をさらに合波することが可能となる。
以上により、図２に示した構成に比して半分の数の可変減衰器２１Ｃ−ｋを設ければよく、その結果、可変減衰器２１Ｃ−ｋの制御の簡素化（つまり、光送信パワーの制御の簡素化）を図ることができる。従って、光合波部２１の大幅な小型化、ひいては、送信局２の大幅な小型化が可能である。
なお、この場合も、図９中に○印や△印を付して示すように、変調器２１Ｂ−（２ｋ−１）から光結合器２１Ｇ−ｋまでの光路長Ｌ_２ｋ−１′と、変調器２１Ｂ−２ｋから光結合器２１Ｇ−ｋまでの光路長Ｌ_２ｋ′とが同じとなるように設計する。
これにより、この場合も、信号Ｑｋとその反転信号Ｑｋバーとの組を互いに位相同期した状態で光合波器２１Ｄにて合波して送信することができ、「チャンネル間クロストーク」の抑圧効果を最大限に発揮させることができる。特に、この場合は、同じ光路長にすべき変調器２１Ｂ−（２ｋ−１），２１Ｂ−２ｋと光結合器２１Ｇ−ｋとの間の距離（光パス）が短くなるので、信号ＱｋとＱｋバーとを位相同期させやすく、設計も容易である。
なお、上記の構成は、例えば図１０に示すように、多重化する波長にそれぞれ別々の信号を載せて伝送する通常のＷＤＭ光伝送システムにおける送信局（光合波部２１′）に適用してもよい。
即ち、通常のＷＤＭ光伝送システムにいても、隣接波長λ_２ｋ−１，λ_２ｋでは伝送損失の違いは軽微であるため、光源２１Ａ−ｉからの光信号を、それぞれ、対応する変調器２１Ｂ−ｉにおいてそれぞれ異なる信号（送信データ）Ｑ_１〜Ｑｎにより変調する構成の場合にも、変調器２１Ｂ−（２ｋ−１）と変調器２１Ｂ−２ｋとの組毎に光結合器２１Ｇ−ｋを設けて、変調器２１Ｂ−（２ｋ−１）の出力と変調器２１Ｂ−２ｋの出力とを光結合器２１Ｇ−ｋにて結合する構成を採る。
これにより、通常のＷＤＭ光伝送システムの送信局に適用される光合波部２１′においても、半分の数の可変減衰器２１Ｃ−ｋで、複数チャンネル（波長）の光送信パワーをチャンネル別ではなく隣接波長λ_２ｋ−１，λ_２ｋ毎に一括制御することができる。
従って、この場合も、可変減衰器２１Ｃ−ｋの台数が節約できるとともに、各可変減衰器２１Ｃ−ｋを制御する回路の縮小化を図ることができるため、総合的に光合波部２１′の低コスト化及び小型化が可能となり、その安定性も向上する。また、この場合も、変調器２１Ｂ−ｉから光合波器２１Ｄまでの光パスが短くなるため、隣接波長λ_２ｋ−１，λ_２ｋ同士の位相同期を保ったままで合波することが容易になる。
（Ｆ）光合波部２１の第２変形例の説明
図２（あるいは図９）により前述した光合波部２１（光信号生成手段２０）は、例えば図１１に示すような構成にしてもよい。
即ち、送信すべき信号Ｑｋをシリアル／パラレル変換してその信号速度（例えば、１０Ｇｂｐｓ）を１／２（５Ｇｂｐｓ）に落とすシリアル／パラレル（Ｓ／Ｐ）変換部２１６と、このＳ／Ｐ変換部２１６の出力（半分）と信号Ｑｋとのいずれか一方を選択して変調器２１Ｂ−（２ｋ−１）の変調信号として出力するセレクタ２１７と、Ｓ／Ｐ変換部２１６の出力（もう半分）と前記の反転ゲート２１Ｅの出力とのいずれか一方を選択して変調器２１Ｂ−２ｋの変調信号として出力するセレクタ２１８とを、隣接波長λ_２ｋ−１，λ_２ｋの組毎に複数設けても良い。
つまり、上記のＳ／Ｐ変換部２１６は、信号Ｑｋの伝送速度変換を行なう伝送速度変換部として機能し、セレクタ２１７，２１８は、信号Ｑｋと反転ゲート２１Ｅの出力との組もしくはＳ／Ｐ変換部２１６の出力のいずれか一方を選択して上記の各変調器２１Ｂ−（２ｋ−１），２１Ｂ−２ｋに入力する選択部として機能するのである。なお、上記の各セレクタ２１７，２１８は、例えば、外部設定などにより選択すべき信号の設定が行なわれる。
上述のごとく構成された光合波部２１では、要求される伝送帯域や光ファイバ伝送路５−１，５−２の条件に応じてセレクタ２１７，２１８の出力を切り替えることにより、信号Ｑｋとその反転信号Ｑｋバーとを１０Ｇｂｐｓのまま２種類の波長λ_２ｋ−１，λ_２ｋで送信する「クロストーク抑圧モード」と、信号Ｑｋのみを５Ｇｂｐｓに落として２種類の波長λ_２ｋ−１，λ_２ｋで送信する「速度変換モード」とを切り替えることができる。
これにより、光ファイバ伝送路５−１，５−２の分散が大きく波形の劣化が抑えにくいとか、非線形性が大きくて高速の信号を伝送することが難しいような場合には、後者の「速度変換モード」を用い、長い伝送距離（中継距離）に対応するためにラマン増幅を使用する場合には、前者の「クロストーク抑圧モード」を用いるといったように、様々な光伝送路の特性や当初導入した構成からアップグレードするといった顧客要求に応えられる付加価値の高い装置（送信局２）を提供することができる。
なお、送信局２の光合波部２１を上記の構成とした場合は、受信局４の光分波部４４においても、上記の「クロストーク抑圧モード」及び「速度変換モード」のいずれかを選択できる構成とする。その詳細については図１６により後述する。
（Ｇ）受信局４の光分波部４４の説明
次に、図１４は受信局４における光分波部４４の構成を示すブロック図であるが、この図１４に示す光合分波部４４は、光分波器４４Ａ，帯域通過フィルタ（ＢＰＦ：Ｂａｎｄ　Ｐａｓｓ　Ｆｉｌｔｅｒ）４４Ｂ−１〜４４Ｂ−ｎ，光受信器４４Ｃ−１〜４４Ｃ−ｎ，特性監視部４４Ｄ−ｋ（ｋ＝１〜ｎ／２），反転ゲート４４Ｅ−ｋ及びセレクタ（ＳＥＬ）４４Ｆ−ｋをそなえて構成されている。
ここで、上記の光分波器４４Ａは、ＥＤＦＡ４３により前置増幅された光ファイバ伝送路５−２からの光信号（ＷＤＭ信号）を各波長λ１〜λｎの光信号に分波するためのもので、例えば、アレイ導波路格子型フィルタが適用される。さらに、上記のＢＰＦ４４Ｂ−ｉは、それぞれ、波長λｉ成分の光信号のみを通過させて雑音成分などの不要成分を除去するためのものであり、上記の光受信器４４Ｃ−ｉは、それぞれ、対応するＢＰＦ４４Ｂ−ｉからの光信号についての受信処理（光電変換など）を施すためのものである。
また、上記の特性（品質）監視部４４Ｄ−ｋは、それぞれ、光受信器４４Ｃ−（２ｋ−１）で受信された波長λ_２ｋ−１の電気信号（信号Ｑ）と光受信器４４Ｃ−２ｋで受信された波長λ_２ｋの電気信号（信号Ｑの反転信号Ｑｋバー）の各特性〔波形や信号誤り率（ビットエラーレート）等〕を監視（モニタ）することにより各波長λ_２ｋ−１，λ_２ｋの信号品質を監視するためのものであり、反転ゲート４４Ｅ−ｋは、それぞれ、光受信器４４Ｃ−２ｋで受信された信号Ｑｋバーを反転して元の信号Ｑｋを得るためのものである。
そして、上記のセレクタ４４Ｆ−ｋは、それぞれ、光受信器４４Ｃ−（２ｋ−１）からの波長λ_２ｋ−１の信号Ｑと、反転ゲート４４Ｅ−ｋからの波長λ_２ｋの信号Ｑｋバーとのいずれか一方を選択するもので、本実施形態では、特性監視部４４Ｄ−ｋでの監視結果に応じた選択制御信号により、信号品質の良い方の信号を受信信号として選択するようになっている。
上述のごとく構成された本実施形態の光分波部４４では、分波４４ＡによりＥＤＦＡ４３からのＷＤＭ信号は、各波長λ１〜λｎの光信号に分波されたのち、ＢＰＦ４４Ｂ−ｉにてそれぞれの雑音成分などの不要成分が除去されて光受信器４４Ｃ−ｉで受信されて電気信号に変換される。
このとき、特性監視部４４Ｄ−ｋでは、それぞれ、光受信器４４Ｃ−（２ｋ−１）で受信された波長λ_２ｋ−１の信号Ｑｋと光受信器４４Ｃ−２ｋで受信された波長λ_２ｋの反転信号Ｑｋバーの各信号品質を、ビットエラーレートの計算などによって監視し、信号品質の良い方が選択されるようセレクタ４４Ｆ−ｋを制御する。
これにより、品質の良い波長λ_２ｋ−１又はλ_２ｋの信号が現用回線（チャンネル）の信号として選択されることになる。
このように、本実施形態の受信局４（光分波部４４）では、送信局２が複数の波長λ_２ｋ−１，λ_２ｋを使って同じ情報内容の信号を伝送していることを活かして、より良い信号品質で受信された方の信号を現用チャンネルの信号として選ぶので、より良い伝送特性を保証することができる。
また、送信局２においてたとえ一部の波長λｉ用の光源２１Ａ−ｉに故障などの異常が発生して、受信局４においてその一部の波長λｉの受信電力が低下してしまうような場合でも、対を成す他の波長によって正常な受信を行なうことができるので、回線（チャンネル）の２重化に近い安全性及び信頼性を得ることができる。
（Ｈ）光分波部４４の第１変形例の説明
次に、図１５は上記の光分波部４４の第１変形例を示すブロック図で、この図１５に示す光分波部４４は、図１４により上述したものと同様の光分波器４４Ａ，ＢＰＦ４４Ｂ−１〜４４Ｂ−ｎおよび光受信器４４Ｃ−１〜４４Ｃ−ｎをそなえるほか、差動増幅器４４Ｇ−ｋ（ｋ＝１〜ｎ／２）をそなえて構成されている。
ここで、これらの差動増幅器４４Ｇ−ｋは、それぞれ、光受信器４４Ｃ−（２ｋ−１）からの電気信号（Ｑｋ）と光受信器４４Ｃ−２ｋからの電気信号（反転信号Ｑｋバー）とを入力とし、これらの差分を検出することで、電気信号の伝送路における同相ノイズ除去用の差動増幅器の原理と同様に、伝送路雑音の直流成分を相殺（キャンセル）しうるものである。
このような構成により、本光分波部４４では、光ファイバ伝送路５−１，５−２で発生するＡＳＥ（Ａｍｐｌｉｆｉｅｄ　Ｓｐｏｎｔａｎｅｏｕｓ　Ｅｍｉｓｓｉｏｎ）等の同相の雑音成分を差動増幅器４４Ｇ−ｋによりキャンセルすることができるため、より良い信号対雑音比を実現することができ、より長い中継距離に対応することが可能となる。
（Ｉ）光分波部４４の第２変形例の説明
図１６は上記の光分波部４４の第２変形例を示すブロック図で、この図１６に示す光分波部４４は、図１１により前述したように送信局２の光合波部２１に、「クロストーク抑圧モード」と「速度変換モード」との切り替え機能を付加した場合の受信側に相当し、上述したものとそれぞれ同様の光分波器４４Ａ，各波長λ１〜λｎ毎のＢＰＦ４４Ｂ−１〜４４Ｂ−ｎ，各波長λ１〜λｎ毎の光受信器４４Ｃ−１〜４４Ｃ−ｎをそなえるほか、反転波受信回路４４１，パラレル／シリアル（Ｐ／Ｓ）変換部４４２及びセレクタ４４３が、送信側の構成（図１１参照）に応じて設けられている。
ここで、上記の反転波受信回路４４１は、例えば、図１４により前述した特性監視部４４Ｄ−ｋ，反転ゲート４４Ｅ−ｋ及びセレクタ４４Ｆ−ｋを含む回路、あるいは、図１５により前述した差動増幅器４４Ｇ−ｋに相当する回路で、光受信器４４Ｃ−（２ｋ−１）の出力と光受信器４４Ｃ−（２ｋ−１）の出力とを入力として受けるようになっている。
従って、送信局２側が「クロストーク抑圧モード」に設定されていれば、本反転波受信回路４４１には、波長λ_２ｋ−１で送られてきた信号Ｑｋと波長λ_２ｋで送られてきた反転信号Ｑｋバーとが入力されることになり、「速度変換モード」が設定されていれば、送信局２側のＳ／Ｐ変換部２１６で速度変換されて（半分に落とされて）２波長λ_２ｋ−１，λ_２ｋを使って送信されてきた信号Ｑｋが入力されることになる。
また、Ｐ／Ｓ変換部４４２は、光受信器４４Ｃ−（２ｋ−１）の出力と光受信器４４Ｃ−（２ｋ−１）の出力とを入力として受けて、その入力信号についてＰ／Ｓ変換（速度変換）を施すためのもので、送信局２側のＳ／Ｐ変換部２１６による速度変換に応じた速度変換（送信側で半分に落とした場合は２倍）を行なうためのものである。
そして、セレクタ４４３は、送信局２側のモード設定に応じたモードが設定されて、その設定に従って上記の反転波受信回路４４１の出力とＰ／Ｓ変換部４４２の出力とのいずれか一方を選択するためのもので、例えば、「クロストーク抑圧モード」であれば反転波受信回路４４１の出力を選択し、「速度変換モード」であればＰ／Ｓ変換部４４２の出力を選択するようになっている。
上述のごとく構成された光分波部４４では、「クロストーク抑圧モード」においては、反転波受信回路４４１の出力が有効となり、波長λ_２ｋ−１で送られてきた信号Ｑｋと波長λ_２ｋで送られてきた反転信号Ｑｋバーとのうち信号品質の良い方あるいは差動増幅器４４Ｇ−ｋによる差分検出結果が出力され、「速度変換モード」においては、Ｐ／Ｓ変換部４４２の出力が有効となり、送信局２側で速度が落とされて（例えば、５Ｇｂｐｓ）２波長λ_２ｋ−１，λ_２ｋを使って送信されてきた信号Ｑｋがその速度を上げて（例えば、１０Ｇｂｐｓ）出力される。
このようにして、送信局２側でのモード設定に応じて動作することにより、送信側と同様に、様々な光伝送路の特性や当初導入した構成からアップグレードするといった顧客要求に応えられる付加価値の高い装置（受信局２）を提供することができる。
（Ｊ）その他
ところで、図１８（Ａ）に模式的に示すように多段の光増幅中継を行なう場合には、光ファイバ伝送路５のもつ波長分散特性のために、上記の信号Ｑｋと波長の異なる反転信号Ｑｋバーとの間には、図１８（Ｂ）に模式的に示すように、伝送（中継）距離が伸びるに従って遅延が累積されることになる。一方、「前方向励起」によるラマン増幅効果は図１８（Ｄ）により前述したように、送信局２および中継局３の出力直後（送信端）で最も大きく得られる。
従って、中継局３においても「チャンネル間クロストーク」を有効に抑圧するためには、中継局３でも信号Ｑｋとその反転信号Ｑｋバーとの間の遅延を補償することが望ましい。そこで、例えば図１７に示すように、少なくとも、中継局２に、その前の光ファイバ伝送路５のもつ波長分散特性を補償するような分散値をもった分散補償器としての分散補償ファイバ（ＤＣＦ：Ｄｉｓｐｅｒｓｉｏｎ　Ｃｏｍｐｅｎｓａｔｉｎｇ　Ｆｉｂｅｒ）３０４を設ける。なお、このＤＣＦ３０４は、入力光パワーに制限がある（入力光パワーが大き過ぎるとノイズ成分が大きくなる）ため、通常、ＥＤＦ３０１の前段に設けられる。
これにより、図１８（Ｃ）に模式的に示すように、信号Ｑｋとその反転信号Ｑｋバーとの間の遅延を中継局３の出力直後で小さくすることができる。この結果多段の光増幅中継を行なうシステム１においても、中継局３に、ＤＣＦ３０４を設けるだけで、伝送距離全体にわたって「チャンネル間クロストーク」の抑圧効果を有効に引き出すことが可能となる。
ただし、「ラマン増幅」は数ｋｍから数１０ｋｍなどという非常に長い距離の光ファイバ伝送路５自体を増幅媒体として用いるため、光ファイバ伝送路５の分散特性や損失特性によりクロストークの発生の仕方が異なり、必ずしも、信号Ｑｋとその反転信号Ｑｋバーを上述したごとく完全に同期した状態（遅延差がゼロの状態）で送信した場合に伝送特性が最良になるとは限らないかも知れない。
そこで、例えば図１９及び図７に模式的に示すように、光の反転信号Ｑｋバー（信号Ｑｋでもよい）の通るパス（誘電体光導波路など）に電極２２１を設け、この電極２２１に屈折率制御回路（タイミング制御回路）２２２から電圧を印可することで光の屈折率を制御して、反転信号Ｑｋバー（信号Ｑｋでもよい）の光路長を調整できるようにしても良い。
これにより、信号Ｑｋとその反転信号Ｑｋバーとの間の遅延差Δτ、即ち、信号Ｑｋとその反転信号Ｑｋバーとの出力タイミングを、適宜に調整することができる。従って、システム運用開始後であっても、温度変化や経年変化に起因するものも含めて上記遅延差Δτを調節することにより、伝送特性を最適化することができ、常に、クロストーク抑圧効果を最大限に発揮させることができる。
なお、上述した実施形態では、隣接する２波長λ_２ｋ−１，λ_２ｋを使って同じ情報内容の信号Ｑｋ（Ｑｋバー）を送信することにより、クロストークを抑圧する場合について説明したが、３波長以上を使っても、上述した実施形態と同様にクロストーク抑圧効果を得ることができる。
例えば、３波長の場合を例にすると、図２０（Ａ）及び図２０（Ｂ）に示すように、信号Ｑｋは波長λ_２ｋ、反転信号Ｑｋバーは波長λ_２ｋ−１と波長λ_２ｋ＋１とを使ってそれぞれ信号Ｑｋの半分のレベル（パワー）で送信する。これにより、この場合も、隣接する３波長λ_２ｋ−１，λ_２ｋ，λ_２ｋ＋１の各光信号を、同期関係を保ったまま波長多重して伝送すればその合計光パワーが一定になるので、ラマン励起光に対する変調効果が抑圧されて、クロストークを抑圧することができる。
さらに、上述した実施形態では、一貫して、「ラマン増幅」時のクロストークの抑圧について説明したが、半導体光増幅器を用いた場合も、上述した実施形態と同様の作用効果が得られる。
即ち、信号Ｑｋとその反転信号Ｑｋバーとを同期した状態で半導体光増幅器に入力すれば、各信号Ｑｋ，Ｑｋバーの合計パワーが一定になるので、半導体光増幅器における活性領域内のキャリア密度の変動が抑えられて、結果として、利得の変動および「パターン効果」による信号波形劣化も抑圧されて、クロストークを有効に抑圧することができるのである。
また、上記の反転信号Ｑｋバーは、必ずしも、信号Ｑｋの完全な反転波形になっていなくても良い。即ち、例えば、反転信号Ｑｋバーと信号Ｑｋとが多少異なる光パワーや波形ずれを有していたとしても、全体としてそれらの合計パワーはほぼ一定になるので、十分、クロストーク抑圧効果を得ることができるものと考えられる。
さらに、上述した実施形態では、光源２１Ａ−ｉからの光信号を信号ＱｋやＱｋバーで外部から変調する外部変調方式を採用しているが、光源２１Ａ−ｉに信号ＱｋやＱｋバーを直接入力して変調をかける直接変調方式を採用しても良い。
また、上述した実施形態では、本発明を、ＥＤＦＡ２２（３３，４３）とラマン増幅器（あるいは半導体光増幅器）とを組み合わせたハイブリッドシステムに適用した場合について説明したが、ラマン増幅器（あるいは半導体光増幅器）単体を用いたＷＤＭ光伝送システムに適用しても、上記と同様の作用効果が得られる。
さらに、上述した実施形態では、信号Ｑｋとその反転信号Ｑｋバーとを隣接する波長λ_２ｋ−１，λ_２ｋを使って伝送しているが、必ずしも隣接波長λ_２ｋ−１，λ_２ｋを使わなくても良い場合もある。例えば、ラマン増幅器の代わりに半導体光増幅器を使う場合には、光信号が増幅を受ける活性領域が数百μｍ〜１ｍｍ程度であるため、光ファイバ伝送路５を増幅媒体として使う場合のような波長分散による遅延の影響はほとんど無いと考えられる。このため、半導体光増幅器を用いた場合には、必ずしも、隣接波長を使う必要は無く、利得帯域内の任意の波長を使用できると考えられる。
また、上述した実施形態では、ＷＤＭ光伝送システム１に、「双方向励起」のラマン増幅を適用した場合について説明したが、勿論、「前方向励起」のみを適用した場合にも、上記と同様の作用効果が得られる。また、上述した実施形態では、送信すべき全ての信号Ｑｋをその反転信号Ｑｋバーとの組で送信するようになっているが、一部の信号Ｑｋのみをその反転信号Ｑｋバーとの組で送信するようにしてもよい。
例えば、一部の信号Ｑｋのみをその反転信号Ｑｋバーとの組で送信するだけで、十分な信号品質で所定距離を伝送できるのなら、残りの信号Ｑｋについては反転信号Ｑｋバーを用いずに通常通りの送信を行なうようにしても良い。また、光伝送路や光増幅器の波長依存性の損失特性により他波長（チャンネル）にクロストークの影響を与えやすい光パワーをもつことになる波長については、反転信号Ｑｋバーとの組で送信し、それ以外の波長については反転信号Ｑｋバーを用いずに送信するようにしてもよい。
このようにすれば、光伝送路や光増幅器の波長依存性の損失特性により波長毎に光パワーのばらつきが生じても、それによるクロストークの影響を抑圧することができる。
そして、本発明は、上述した実施形態に限定されるものではなく、上記以外にも、本発明の趣旨を逸脱しない範囲で種々変形して実施することができる。
産業上の利用可能性
以上のように、本発明によれば、波長多重光伝送システムにおいて、「前方向励起」のラマン増幅を用いた場合に顕著に現われるチャンネル間クロストークを光デバイスの性能や特性に依存せずに効果的に抑圧することができるので、波長多重光信号を従来よりも低雑音で長距離伝送することが可能となり、その有用性は極めて高いものと考えられる。
【図面の簡単な説明】
図１は本発明の一実施形態に係る波長多重（ＷＤＭ）光中継伝送システムの構成を示すブロック図である。
図２は図１に示す送信局における光合波部の構成を示すブロック図である。
図３は図２に示す光源及び変調器の部分に着目した構成を示すブロック図である。
図４（Ａ）は本実施形態に係るラマン励起光と送信すべき信号及びその反転信号とについての波長（チャンネル）配置例を示す模式図である。
図４（Ｂ）は本実施形態に係るラマン増幅前のラマン励起光，送信すべき信号及びその反転信号の波形例を示す模式図である。
図４（Ｃ）は本実施形態に係るラマン増幅後のラマン励起光，送信すべき信号及びその反転信号の波形例を示す模式図である。
図５は本実施形態に係る反転信号生成方法の第１変形例を説明するためのブロック図である。
図６は図５に示す変調器に対するバイアス制御方法を説明するための模式図である。
図７は本実施形態に係る反転信号生成方法の第２変形例を説明するためのブロック図である。
図８は本実施形態に係る反転信号生成方法の第３変形例を説明するためのブロック図である。
図９は図１及び図２に示す光合波部の第１変形例を示すブロック図である。
図１０は図９に示す光合波部が通常のＷＤＭ光伝送システムに適用できることを説明するためのブロック図である。
図１１は図１及び図２に示す光合波部の第２変形例を示すブロック図である。
図１２は図１に示すＥＤＦＡの構成を示すブロック図である。
図１３は図９（あるいは図１０）に示す光合波器の通過帯域特性例を示す模式図である。
図１４は図１に示す受信局における光分波部の構成を示すブロック図である。
図１５は図１に示す光分波部の第１変形例を示すブロック図である。
図１６は図１に示す光分波部の第２変形例を示すブロック図である。
図１７は図１に示すＥＤＦＡの変形例を示すブロック図である。
図１８（Ａ）は多段の光増幅中継を行なう場合のＷＤＭ光伝送システムを示すブロック図である。
図１８（Ｂ）は図１８（Ａ）に示すシステムにおいてＤＣＦを設けない場合の送信信号とその反転信号との伝送距離に応じた遅延差を表わす模式図である。
図１８（Ｃ）は図１８（Ａ）に示すシステムにおいて中継局にＤＣＦを設けた場合の送信信号とその反転信号との伝送距離に応じた遅延差を表わす模式図である。
図１８（Ｄ）は図１８（Ａ）に示すシステムにおける「前方向励起」のラマン増幅による伝送距離に応じたラマン利得を表わす模式図である。
図１９は本実施形態に係る送信すべき信号とその反転信号との遅延差制御を説明するための模式図である。
図２０（Ａ）及び図２０（Ｂ）はいずれも本実施形態に係る送信すべき信号とその反転信号との送信に３波長を使う場合を説明するための模式図である。
図２１は従来のＥＤＦＡとラマン増幅器とを併用したＷＤＭ光伝送システムの一例を示すブロック図である。
図２２は従来のＥＤＦＡとラマン増幅器とを併用したＷＤＭ光伝送システムにおける中継利得及び自然放出光雑音を説明するための模式図である。
図２３（Ａ）及び図２３（Ｂ）はいずれもラマン増幅時のラマン励起光に対する変調効果を説明するための模式図である。
図２４（Ａ）はラマン励起光と送信すべき２つの信号との波長（チャンネル）配置例を示す模式図である。
図２４（Ｂ）は図２４（Ａ）に示すラマン励起光と送信すべき２つの信号光のラマン増幅前の波形例を示す模式図である。
図２４（Ｃ）は図２４（Ｂ）に示すラマン励起光と送信すべき２つの信号光のラマン増幅後の波形例を示す模式図である。
図２５（Ａ）は「前方向励起」のラマン増幅器構成を示すブロック図である。
図２５（Ｂ）は「後方向励起」のラマン増幅器構成を示すブロック図である。
図２５（Ｃ）は「双方向励起」のラマン増幅器構成を示すブロック図である。
図２６（Ａ）〜図２６（Ｃ）はいずれも半導体光増幅器の「パターン効果」を説明するための模式図である。
図２７（Ａ）〜図２７（Ｅ）はいずれも半導体光増幅器の「パターン効果」による「チャンネル間クロストーク」を説明するための模式図である。Technical field
The present invention relates to an optical transmitter, an optical repeater, an optical receiver, and an optical transmission method, and more particularly to a wavelength for collectively compensating for transmission loss of an optical signal (wavelength multiplexed optical signal) in an optical fiber transmission line using an optical amplifier. The present invention relates to an optical transmitter, an optical repeater, an optical receiver, and an optical transmission method suitable for use in a multiplex optical transmission system.
Background art
2. Description of the Related Art In recent years, in order to cope with a demand for information communication which is rapidly increasing with the spread of the Internet, a wavelength division multiplexing (WDM) optical transmission technology using a rare earth-doped optical fiber amplifier such as an EDFA (Erbium Doped Fiber Amplifier) has been put to practical use. , Is being introduced. However, in order to cope with a communication network centered on data traffic in the future, it is necessary to provide a broadband information communication system which is far higher than before at a low cost.
Therefore, it is necessary to increase the wavelength multiplexing density and increase the number of wavelength multiplexes by expanding the amplification wavelength band of the fiber-type optical amplifier, and to reduce the cost of the entire system by increasing the relay interval. Have been. Expectations are also being raised for the realization of photonic networks that perform switching and routing in the optical domain.
For that purpose, it is required to reduce the cost, size, and power consumption of the optical amplifier. Currently, optical amplifiers include Raman amplifiers and semiconductor optical amplifiers, in addition to rare earth-doped optical fiber amplifiers such as EDFAs. By utilizing the features of these optical amplifiers, it is expected that an optical amplifier that fulfills the role of complementing the rare-earth-doped optical fiber amplifier and satisfies the above-described requirements will be realized.
For example, a Raman amplifier has attracted attention as one of the methods for widening the amplification (gain) band of an optical amplifier. Rare-earth-doped optical fiber amplifiers such as EDFAs amplify using the transition between the levels of rare-earth atoms added to the optical fiber, so the band in which optical amplification can be performed depends on the type of added atoms. For example, in the case of EDFA, it is limited to about 1530 to 1600 nm (nanometer).
On the other hand, a Raman amplifier performs amplification using the "stimulated Raman scattering phenomenon" generated in an optical fiber, and therefore has an amplification characteristic in which a gain peak occurs on the longer wavelength side (about 100 nm) of an excitation wavelength. ing. Therefore, optical amplification can be performed in an arbitrary wavelength band by selecting the excitation light wavelength. Therefore, if a Raman amplifier and a rare earth-doped optical fiber amplifier such as an EDFA are connected in series, it is possible to widen the gain band.
In addition, the above-mentioned “stimulated Raman scattering phenomenon” means that when a high power light is input to an optical fiber, a part of the input light power is consumed by the lattice vibration in the optical fiber, so that a part of the input light is Utilizes a "Raman scattering phenomenon" that is converted into light (called Stokes light or natural Raman scattered light) on a longer wavelength side than the wavelength of the input light. It takes advantage of the fact that the conversion occurs remarkably.
In addition, in Raman amplification, since a superimposed gain can be obtained by using pump lights of a plurality of wavelengths, a method of using this to widen the gain band has been proposed (for example, a special method). See Japanese Unexamined Patent Publication No. 10-73852). Further, in Raman amplification, since an optical fiber transmission line itself is used as an amplification medium, amplification of an optical signal is performed in a distributed manner. Therefore, in Raman amplification, amplification with lower noise is possible than in the case of using a rare-earth-doped optical fiber amplifier having the same gain in which lumped constant amplification is performed (reference: Nonlinear Fiber Optics, published by Academic Press).
Therefore, for example, as described in JP-A-10-22931, the transmission distance of an optical signal can be extended by combining a Raman amplifier and a rare earth-doped optical fiber amplifier such as an EDFA. That is, as shown in FIG. 21, a rare earth-doped optical fiber 111, a pumping light source 112 for the rare-earth-doped optical fiber 111, a rare-earth-doped optical fiber amplifier 110 having a multiplexing coupler 113, a pump light source 121 for Raman amplification, and An optical repeater 100 having a multiplex coupler 122 for inputting the optical output to the optical fiber transmission line 101 is provided in a WDM optical transmission system. In FIG. 21, reference numeral 102 denotes an optical fiber transmission line for transmitting the optical output after amplification by the rare earth-doped optical fiber amplifier 110.
Here, the pump light source 121 has a wavelength suitable for causing Raman amplification (stimulated Raman scattering phenomenon) at an optical signal wavelength in the optical fiber transmission line 101, and an optical output level sufficient to realize a predetermined gain. The pumping light is transmitted through the multiplex coupler 22 to the optical fiber transmission line 101 (the direction opposite to the traveling direction of the optical signal).
As a result, an stimulated Raman scattering phenomenon occurs in the optical fiber transmission line 101 and an optical signal (hereinafter, also referred to as signal light) propagating through the optical fiber transmission line 101 is Raman-amplified, and an optical signal input to the optical repeater 100 Are amplified to a predetermined level. Therefore, in the optical repeater 100, the gain (relay gain) of the rare-earth-doped optical fiber amplifier 110 required to obtain the same optical output level as when Raman amplification is not used is reduced. As a result, the amplification gain of the rare-earth-doped optical fiber amplifier 110 has a margin, and the transmission distance of the optical signal can be extended within a range in which the noise effect due to the amplification is allowed.
Therefore, for example, as shown in FIG. 22, when a WDM optical transmission system is configured by a plurality of optical repeaters 100, the optical signal transmitted from the optical transmitter 130 is transmitted to the optical receiver 100 while being relayed by each optical repeater 100. When the light is transmitted to the optical fiber 140, the light level decreases each time the light passes through the optical fiber transmission line 101 (102). The optical level increases as compared with the case where Raman amplification is not used (see the broken line 300), and the relay gain required for the rare-earth-doped optical fiber amplifier 110 decreases. At this time, as shown in FIG. 22, the spontaneous emission optical noise can be suppressed (see the solid line 400) as compared with the case where the Raman amplification is not used (see the broken line 500).
As a result, the relay distance between the optical repeaters 100 can be increased as compared with the case where Raman amplification is not used, and a smaller number of optical repeaters is required to configure a WDM optical transmission system having the same transmission distance. In other words, it is possible to construct a system at low cost.
As an optical amplifier for realizing a photonic network, a semiconductor optical amplifier that has advantages of being smaller and consuming less power than fiber-type optical amplifiers such as rare-earth-doped optical fiber amplifiers is expected. Since this semiconductor optical amplifier has a high-speed switching characteristic unlike a fiber-type optical amplifier such as an EDFA, it is particularly expected to be applied as an optical gate element in an optical cross-connect (Reference: S. Araki, et. al "A 2.56 Tb / s Throughput Packet / Cell-Based Optical Switch-Fabric Demonstrator", Technical Digest of ECOC'98, vol.3 p.127).
Further, since the semiconductor optical amplifier is based on a semiconductor, it has an advantage that it can be realized as a multi-channel array module by hybrid integration with a quartz-based planar optical circuit.
As described above, the Raman amplifier and the semiconductor optical amplifier, which are promising next-generation optical amplifiers, have wavelengths (both due to their response speeds that are much faster than fiber-type optical amplifiers such as EDFAs). There has been a problem that crosstalk between channels has occurred, which has not been a problem with fiber-type optical amplifiers.
For example, an EDFA has a relatively slow relaxation time of erbium atoms and a response speed on the order of milliseconds. Therefore, when a modulated optical signal on the order of Gbps (gigabits per second) is input, only the average light intensity is felt. No, the waveform of the optical signal is not distorted. On the other hand, the stimulated Raman scattering effect in the optical fiber, which is the basis of the Raman amplification, is a nonlinear interaction of signal lights of all wavelengths propagating in the optical fiber, so that the response speed is as high as about picosecond (ps). It is known that it is early.
For this reason, in the gain saturation state where the input light level is large, for example, as schematically shown in FIGS. 23A and 23B, the energy when amplifying the intensity-modulated signal light Q (wavelength λ1) is increased. A phenomenon occurs in which the intensity of the pumping light P (wavelength λ0), which is deprived of the light, is modulated following (the pumping light P fluctuates). When multi-wavelength batch amplification is performed, the fluctuation of the pump light P is converted into the amplification degree of another signal light, thereby causing crosstalk between wavelengths (channels).
For example, as schematically shown in FIGS. 24 (A) and 24 (B), the signal light Q of wavelength λ1 (bit pattern = 101) and the signal light Q ′ of wavelength λ2 (bit pattern = 111) are pumped light. When collectively amplifying by P (wavelength λ0), as schematically shown in FIG. 24C, the intensity of the pump light P when collectively amplifying the same bit value (“1, 1”) (see reference numeral 201). And the intensity (see reference numeral 202) of the pump light P when the different bit values (1, 0) are collectively amplified is different from each other, so that the pump light P fluctuates.
Then, in this case, the intensity (see reference numeral 202) of the pump light P when the different bit values (1, 0) are collectively amplified is converted into the amplification degree of the signal light Q ′ having the bit value “1” to be amplified. As a result, the waveform of the signal light Q ′ is distorted from the original waveform, as indicated by reference numeral 203 in FIG. This is the "crosstalk between channels" phenomenon.
Here, in the configuration method of the Raman amplifier, for example, as shown in FIG. 25A, a Raman pumping light source 121 and a multiplexer 122 are provided in a stage preceding the optical fiber transmission line 103, and the same direction as the propagation direction of the signal light is provided. 25A. On the other hand, as shown in FIG. 25B, a Raman pumping light source 121 and a multiplexer 122 are provided at the subsequent stage of the optical fiber transmission line 103, and the propagation direction of the signal light. The pumping light is input in the opposite direction to the "backward pumping" (the one described above with reference to FIG. 21 is also of this type), and these "forward pumping" and "backward pumping" as shown in FIG. And "bidirectional excitation".
Among them, the phenomenon of “crosstalk between channels” has a configuration of “forward pumping” in which the signal light intensity at the pumping light input (multiplexing) point is strong and the signal light and the pumping light propagate in the same direction. [Reference: OPTRONICS (1999) No. 8 (Noboru Egawa), "Bandwidth of crosstalk in Raman amplifiers", OFC'94 Technical Digest (Fabrizio Forghieri, et al.). For this reason, when "forward excitation" is employed, waveform deterioration due to "cross-talk between channels" is a factor that limits the transmission distance.
On the other hand, in the case of “backward pumping” in which the signal light and the pumping light propagate in opposite directions, the influence of “inter-channel crosstalk” (hereinafter, also simply referred to as “crosstalk”) is small, but the signal light and the pumping light Since the excitation light is strong and the signal light is weak at the multiplexing point, natural Raman scattered light is likely to be generated, and noise characteristics are inferior. Therefore, in order to extend the optical repeater interval using a Raman amplifier, the effect of "crosstalk" is suppressed in some way, and optimization is made to maximize the advantages of both by "bidirectional pumping" It is effective.
Note that the influence of "crosstalk" can be avoided by increasing the pumping light intensity because the intensity of the pumping light becomes smaller if the intensity of the pumping light is sufficiently larger than that of the signal light. At present, the pumping light output intensity up to about several hundred mW is the limit depending on the performance of the device, and it becomes difficult to supply sufficient optical power when the number of wavelengths increases. Therefore, it is necessary to suppress "crosstalk" by some method.
By the way, it is known that such a “crosstalk” phenomenon similarly occurs in a semiconductor optical amplifier. That is, since the semiconductor optical amplifier is a device that amplifies incident light by stimulated emission by utilizing population inversion caused by carrier injection into the semiconductor active layer, the carrier density in the active layer depends on the incident light intensity. Change.
For this reason, the carrier relaxation time becomes a problem as in the case of the Raman amplifier. However, since the carrier relaxation time of the semiconductor optical amplifier is on the order of subnanoseconds (reference: Mukai et al., “1.5 μm band InGaAsP / InP resonance”). Type laser amplifier ", IEICE Trans., Vol. J69-C. No. 4, pp 421-431 (1986)), when amplifying input signal light of the order of Gbps, the response is about the same as that of the Raman amplifier. You will have speed.
Accordingly, for example, as schematically shown in FIGS. 26A to 26C, a phenomenon in which a change in carrier density follows a change in signal light occurs. Waveform distortion (called “pattern effect”) depending on the input signal light pattern occurs in light. For this reason, even when multi-wavelength batch amplification is performed by the semiconductor optical amplifier, “inter-channel crosstalk” occurs. This situation is schematically shown in FIGS. 27 (A) to 27 (E).
That is, if the modulated input light "1" (see FIG. 27A) and the DC input light "2" (see FIG. 27B) are input to the semiconductor optical amplifier, respectively, The carrier density also changes in accordance with the change in the total optical power (see FIG. 27C), and the amplification of all channels is modulated accordingly. Therefore, the output light “1” has a waveform distortion as shown in FIG. 27D, and the output light “2” depends on the output light “1” as shown in FIG. 27E. Crosstalk occurs.
Since this phenomenon occurs at the time of gain saturation, it can be avoided by increasing the saturation output of the semiconductor optical amplifier. However, as with the Raman amplifier, there is a limit in device characteristics. It is necessary to suppress.
SUMMARY OF THE INVENTION The present invention has been made in view of the above problems, and reduces crosstalk between channels that occurs when multi-wavelength collective amplification is performed by an optical amplifier such as a Raman amplifier or a semiconductor optical amplifier. It is an object of the present invention to enable effective suppression in a method that does not depend on.
Disclosure of the invention
In order to achieve the above object, an optical transmitter according to the present invention includes: an optical signal generating unit that generates a main signal to be transmitted and an inverted signal thereof as optical signals of a plurality of wavelengths; Wavelength multiplexing means for wavelength-multiplexing and transmitting the optical signals of the plurality of wavelengths generated as described above.
As a result, the main signal to be transmitted and its inverted signal are transmitted as wavelength multiplexed optical signals of a plurality of wavelengths, so that the total optical power of these main signal and its inverted signal becomes substantially constant. Therefore, even if the main signal and its inverted signal are collectively amplified by the pump light, the intensity of the pump light follows the main signal waveform and is modulated as in the related art (the pump light fluctuates). Can be suppressed without depending on the optical device characteristics, and crosstalk between main signals can be suppressed reliably.
Here, the main signal and the inverted signal are preferably output in a synchronized state. In this case, at least these two optical signals are wavelength-multiplexed and transmitted while maintaining the synchronized state. Therefore, the total optical power can be always kept constant at the transmitting end. Therefore, for example, in the “forward pumping” configuration in which the maximum Raman amplification effect is obtained at the transmitting end, a sufficient crosstalk suppression effect can be obtained.
The optical signal generating means includes an inverting circuit for inverting a main signal as an electric signal, a first light source for generating an optical signal of a certain wavelength, and a wavelength of the optical signal generated by the first light source. A second light source that generates an optical signal having a wavelength different from that of the first light source, a first modulator that modulates an optical signal from the first light source with the main signal, and a light from the second light source. A second modulator for modulating a signal with the output of the inverting circuit may be provided.
With this configuration, the main signal to be transmitted is inverted in the state of an electric signal before being input to the modulator, and the optical signal from each of the above light sources is respectively converted into the inverted electric signal and the electric signal before the inversion. By performing the modulation, a main signal and an inverted signal as an optical signal can be obtained. Therefore, in order to obtain an inverted signal of the optical signal, it is sufficient to improve only the electric circuit without making any changes to the optical components of the existing optical transmitter, so that the present optical transmitter can be realized very easily. .
Further, as another mode for obtaining a main signal and an inverted signal as an optical signal, for example, a first light source for generating an optical signal of a certain wavelength, A second light source that generates an optical signal having a wavelength different from the wavelength of the optical signal generated by the first and second light sources, and first and second light sources that respectively modulate the optical signals from these light sources with the same main signal as an electric signal. A modulator and a modulation state for controlling the modulation state of each of the modulators such that a main signal is output as an optical signal from one modulator and an inverted signal is output as an optical signal from the other modulator. It is conceivable to provide a control circuit.
In this way, the main signal and the inverted signal of the optical signal can be obtained only by controlling the modulation state of each of the modulators. Therefore, there is no need for an inverting circuit for inverting an electric signal as described above, and cost reduction and miniaturization are possible. Further, it is possible to avoid occurrence of a delay between the main signal and the inverted signal due to a different path of the electric signal (whether the signal passes through an inversion circuit or not) (that is, the main signal as an optical signal). And the inverted signal can be obtained in a more synchronized state).
Further, as still another mode for obtaining a main signal and an inverted signal of an optical signal, an optical multiplexer for multiplexing a main signal and a DC signal as an optical signal with the optical signal generating means, It is conceivable to provide a semiconductor optical amplifier having the output of the device as an input.
With this configuration, as described above, the intensity of the DC signal is modulated following the main signal waveform due to the crosstalk characteristic inherent to the semiconductor optical amplifier. The main signal as an optical signal and its inverted signal can be obtained.
Therefore, also in this case, there is no need to invert the signal with an electric signal, and only one semiconductor optical amplifier functioning as a modulator is required, so that cost reduction and size reduction can be achieved. Also, there is no delay between the main signal and the inverted signal. Further, in this case, since the input light can be handled as input without converting it into an electric signal, it is possible to configure without using a light source.
The optical path length from the first modulator to the wavelength multiplexing means may be the same as the optical path length from the second modulator to the wavelength multiplexing means. In this way, wavelength multiplexing can be performed without causing a delay difference between the main signal and its inverted signal, so that these two signals can be wavelength multiplexed in a synchronized state and transmitted. And the crosstalk suppressing effect can be maximized.
Here, the optical signal generating means may be provided with a variable attenuator for adjusting the output level of each of the modulators. In this way, the light levels (powers) of the main signal and its inverted signal can be individually adjusted, and the total power of the main signal and its inverted signal is used to maximize the crosstalk suppressing effect. It is possible to control to an optimal state.
Further, the optical signal generation means includes an optical coupler for coupling the outputs of the first and second modulators, and an optical path length from the first modulator to the optical coupler; The optical path length from the second modulator to the optical coupler may be the same.
Even in this case, since the main signal and the inverted signal can be combined without causing a delay difference, these two signals can be transmitted in a reliably synchronized state, The crosstalk suppression effect can be maximized. Further, since the optical path length from the modulator to the optical coupler can be shortened, it is easy to combine the main signal and its inverted signal while maintaining the synchronous relationship, and the design is also easy. .
In this case, if a variable attenuator for adjusting the output level of the optical coupler is provided, a smaller number of variable attenuators can be used than in the case of individually adjusting the output level of each optical modulator. It is possible to control the total power of the main signal and its inverted signal to an optimum state in which the crosstalk suppression effect is maximized.
The wavelength multiplexing means may be configured using an optical multiplexer having the plurality of types of wavelengths as pass bands per channel. In this way, it is possible to further combine optical signals including a plurality of types of wavelengths.
Further, the optical signal generation means includes a transmission rate conversion unit for performing transmission rate conversion of the main signal, a set of the main signal and the output of the inversion circuit, or an output of the transmission rate conversion unit. A selector for selecting one of the modulators and inputting the selected modulator to each of the modulators may be provided.
In this case, for example, when the dispersion is large and it is difficult to suppress the deterioration of the signal waveform, or when the nonlinearity is large and it is difficult to transmit a high-speed signal, the transmission rate is reduced by the above-described transmission rate conversion. If a long transmission distance is to be used, a high-value-added optical transmitter that can respond to various transmission path characteristics and customer requirements, such as using the above-mentioned inverted signal to suppress crosstalk. Can be provided.
By the way, each of the above modulators may be configured as a Mach-Zehnder type optical modulator / multiplexer for multiplexing different output ports of two Mach-Zehnder type optical modulators. In this case, the above modulator can be realized with a very simple configuration and can be realized by being integrated on one substrate, which can reduce the cost and size of the optical transmitter. Contribute greatly.
Further, the optical signal generating means may include a timing control circuit for controlling output timing of the main signal and its inverted signal. In this way, the delay difference between the main signal and its inverted signal can be adjusted as appropriate, so that it is not always necessary to transmit the main signal and its inverted signal in a synchronized state. If the crosstalk suppression effect cannot be said to be maximized, or the change in the delay difference due to temperature change or aging can be adjusted, the crosstalk suppression effect is always maximized. Can be done.
Further, the optical transmitter according to the present invention is provided with a plurality of light sources each generating an optical signal having a different wavelength, and provided for each of these light sources, and modulates the optical signal from each light source with a main signal to be transmitted. A modulator, an optical coupler for coupling the outputs of these modulators at least for every two adjacent wavelengths, a variable attenuator for adjusting the output level of the optical coupler, and an output of the variable attenuator. It is characterized in that it is configured with an optical multiplexer that oscillates.
In other words, the optical transmitter described above pays attention to the fact that the difference in transmission loss between adjacent wavelengths is insignificant. The output level after coupling adjacent wavelengths is controlled by a variable attenuator without providing a device. Therefore, the number of necessary variable attenuators can be saved, and as a result, the circuit for controlling the variable attenuator can be reduced, so that the cost can be reduced overall and the stability can be improved. I do.
Further, the optical repeater of the present invention is for relaying an output of an optical transmitter for transmitting a main signal to be transmitted and an inverted signal thereof as wavelength-multiplexed optical signals of a plurality of wavelengths, and It is characterized by having a dispersion compensator for compensating the chromatic dispersion of the main signal and its inverted signal.
Therefore, according to the optical repeater of the present invention, the delay difference (dispersion) between the main signal and the inverted signal having different wavelengths accumulated as the transmission distance increases due to the chromatic dispersion characteristics of the optical transmission line is calculated as described above. , And the above-described crosstalk suppressing effect can be maximized even in long-distance transmission.
Further, the optical receiver of the present invention receives an output of an optical transmitter for transmitting a main signal to be transmitted and an inverted signal thereof as a wavelength multiplexed optical signal of a plurality of wavelengths, A quality monitoring unit for monitoring the quality of the inverted signal and the inverted signal, and a selecting unit for selecting one of the main signal and the inverted signal as a reception signal according to the quality monitoring result of the quality monitoring unit. It is characterized by having been done.
Therefore, according to the optical receiver of the present invention, it is possible to select, for example, an optical signal having a higher quality wavelength as the working signal in accordance with the quality monitoring result by the above-described quality monitoring unit. In addition to guaranteeing good transmission characteristics, it is possible to obtain reliability close to the redundancy of the line.
Further, the optical receiver of the present invention receives an output of an optical transmitter for transmitting a main signal to be transmitted and an inverted signal thereof as wavelength-multiplexed optical signals of a plurality of wavelengths. An optical demultiplexer for demultiplexing the optical signal into the main signal and the inverted signal, and a differential amplifier that receives the main signal and the inverted signal from the optical demultiplexer as inputs. Features.
Therefore, according to this optical receiver, the DC component of the transmission line noise added to the wavelength-division multiplexed optical signal (main signal and inverted signal) can be canceled (canceled) by the differential amplifier. Ratio can be realized, and a longer transmission distance can be accommodated.
Note that adjacent wavelengths may be used as the plurality of types of wavelengths. In this case, the influence of the optical transmission line having the wavelength-dependent transmission loss is reduced as compared with the case where non-adjacent wavelengths are used. For example, an optical signal of a plurality of wavelengths is converted into an optical signal of one wavelength. Assuming that the transmission power can be controlled collectively, it greatly contributes to the simplification of the optical transmission power control and the miniaturization of the optical transmitter.
BEST MODE FOR CARRYING OUT THE INVENTION
Hereinafter, embodiments of the present invention will be described with reference to the drawings.
(A) Description of one embodiment
FIG. 1 is a block diagram showing a configuration of a wavelength division multiplexing (WDM) optical transmission system according to an embodiment of the present invention. The WDM optical transmission system 1 shown in FIG. A relay station (optical repeater) 3 connected to the transmitting station 2 via an optical (fiber) transmission path 5-1 and a reception station connected to the relay station 3 via an optical (fiber) transmission path 5-2. A station (optical receiver) 4 is provided. In FIG. 1, the relay station 3 has a single configuration. Of course, a plurality of relay stations may or may not be provided depending on the transmission distance.
Then, as shown in FIG. 1, focusing on the main parts, the transmitting station 2 is provided with an optical multiplexing unit 21, an EDFA 22, a Raman excitation light source 23, and an optical multiplexer 24, and the relay station 3 is provided with: The Raman excitation light sources 31, 34, the optical multiplexers 32 and 35, and the EDFA 33 are provided. Further, the receiving station 4 is provided with the Raman excitation light source 41, the optical multiplexer 42, the EDFA 43, and the optical demultiplexer 44.
Here, in the transmitting station 2, the optical multiplexing unit 21 is for generating a WDM signal to be transmitted to the receiving station 4, and the EDFA (rare-earth-doped optical fiber amplifier) 22 is used for transmitting the WDM signal. This is for amplifying a WDM signal in a predetermined wavelength band (for example, 1.55 μm band) with a predetermined amplification gain. For example, as shown in FIG. 12, an EDF (rare-earth-doped optical fiber) 301 and excitation for this EDF 301 are provided. An excitation light source 302 for generating light and an optical multiplexer 303 for inputting the excitation light from the excitation light source 302 to the EDF 301 are provided. The configuration of EDFAs 33 and 43 described later is the same as that shown in FIG.
Further, the Raman pump light source 23 is a pump light (for forward pumping) having a wavelength suitable for performing Raman amplification in the same wavelength band as the EDFA 22 in the optical fiber transmission line 5-1 (hereinafter referred to as Raman pump light). The optical multiplexer 24 is for multiplexing the output of the EDFA 22 and the Raman pump light from the Raman pump light source 24 and outputting the multiplexed light to the optical fiber transmission line 5-1. For example, this can be realized by applying an arrayed waveguide grating filter.
Further, in the relay station 3, the input-side Raman pumping light source 31 has a wavelength (for backward pumping) of a wavelength suitable for performing Raman amplification in the same wavelength band as the EDFA 22 in the optical fiber transmission line 5-1. The optical multiplexer 32 on the input side is for generating pumping light, and is for inputting the Raman pumping light from the Raman pumping light source 31 to the optical fiber transmission line 5-1. This is similar to the EDFA 22 in the transmitting station 2, and is for amplifying the WDM signal from the optical fiber transmission line 5-1 passing through the optical multiplexer 32 with a predetermined amplification gain.
On the other hand, the output side Raman pumping light source 34 supplies Raman pumping light (for forward pumping) having a wavelength suitable for performing Raman amplification in the same wavelength band as the EDFAs 22 and 33 in the optical fiber transmission line 5-2. The optical multiplexer 35 on the output side combines the output of the EDFA 33 and the Raman pumping light from the Raman pumping light source 34 and outputs the combined light to the optical fiber transmission line 5-2. is there.
Further, in the receiving station 4, the Raman pump light source 41 is a Raman pump (for backward pumping) having a wavelength suitable for performing Raman amplification in the same wavelength band as the EDFAs 22 and 33 in the optical fiber transmission line 5-2. The optical multiplexer 42 is for generating light, and is for inputting the Raman pump light from the Raman pump light source 41 to the optical fiber transmission line 5-2.
The EDFA 43 is similar to the EDFAs 22 and 33, and amplifies the WDM signal from the optical fiber transmission line 5-2 passing through the optical multiplexer 43 with a predetermined amplification gain. The demultiplexing unit 44 is for demultiplexing the output (WDM signal) of the EDFA 43 for each wavelength multiplexed and performing a predetermined reception process on the optical signal of each wavelength.
That is, the WDM optical transmission system 1 of the present embodiment (hereinafter, may be simply abbreviated as “system 1”) uses the above-described “bidirectional pumping” in the optical relay transmission system using the EDFAs 22, 33, and 43. It has a hybrid configuration to which Raman amplification is applied.
With such a configuration, in the present system 1, after the WDM signal obtained by the optical multiplexing unit 21 of the transmitting station 2 is commonly amplified by the EDFA 22, the WDM signal is transmitted to the Raman pumping light from the Raman pumping light source 23 and the optical multiplexer 24. Multiplexed and transmitted to the optical fiber transmission line 5-1.
In the relay station 3, in addition to the amplification by the EDFA 33, the Raman pump light is injected from the Raman pump light sources 32 and 34 into each of the optical fiber transmission lines 5-1 and 5-2, thereby the optical fiber transmission line 5. Raman amplification of "bidirectional excitation" using -1, 5-2 as an amplification medium is performed. The Raman amplification at this time is driven under the condition that the gain is in the same wavelength band as the EDFAs 22 and 33.
The receiving station 4 similarly transmits the optical fiber transmission line 5-2 by injecting the Raman excitation light from the Raman excitation light source 41 into the optical fiber transmission line 5-2 by the optical multiplexer 42. Raman amplification of the WDM signal is performed. After that, the WDM signal that has been Raman-amplified and received is pre-amplified by the EDFA 43, and then is demultiplexed and received by the optical demultiplexing unit 44.
In this manner, the WDM signal transmitted from the transmitting station 2 has a reduced optical level due to the transmission loss characteristics of the optical fiber transmission lines 5-1 and 5-2. 2 is used as an amplifying medium, and is Raman-amplified by bidirectional Raman pumping light, so that the input light level to the relay station 3 and the receiving station 4 is much larger than when Raman amplification is not used (“forward pumping” and “ (Compared to the case where only one of the “backward excitation” is applied).
As a result, the amplification gains required for the EDFAs 22, 33, and 43 can be greatly reduced, and the Raman amplification is a distributed constant type amplification and has excellent low noise characteristics as described above. It is possible to greatly increase the relay distance of a WDM signal under transmission conditions.
Here, when Raman amplification by “bidirectional pumping” is performed as described above, that is, when Raman amplification by “forward pumping” needs to be performed, “inter-channel crosstalk” is generated as described above. It becomes a problem.
Therefore, in the present embodiment, as shown in FIG. 2, for example, the optical multiplexing unit 21 is used for each wavelength λ1 to λn (where n is an even number of 2 or more, for example, 16, 32, 64, 128, etc.). 21A-1 to 21A-n, modulators (external modulators) 21B-1 to 21B-n, variable attenuators 21C-1 to 21C-n, and an optical multiplexer 21D. As shown in FIG. 3, two adjacent wavelengths λ _2k-1 , Λ _2k (However, an inverting gate (inverting circuit) 21E is provided for each set of the modulators 21B- (2k-1) and 21B-2k corresponding to k = 1 to n / 2).
Here, each of the light sources 21A-i (i = 1 to n) is for generating an optical signal of the wavelength λi, and is configured by, for example, a semiconductor laser or the like. The wavelengths λ1 to λn are, of course, the wavelength bands included in the amplification wavelength bands of the EDFAs 22, 33, and 43, for example, the 1.55 μm band. In the present embodiment, λ1 is assumed to be on the short wavelength side.
Next, each of the external modulators 21B-i modulates an optical signal (wavelength λi) from the corresponding light source 21-i. Here, as shown in FIG. The optical signal (wavelength λ) from the (first) light source 21A- (2k-1) is output by the external modulator 21B- (2k-1). _2k-1 ) Is modulated by the main signal (transmission data; electrical signal) Qk to be transmitted, and the (second) external modulator 21B-2k outputs an optical signal (wavelength λ) from the (second) light source 21A-2k. _2k ) Is the signal Qk wave at the inversion gate 21E.

). Note that, for example, a known Mach-Zehnder type optical modulator is applied to these external modulators 21B-i as described later.
Further, each variable attenuator 21C-i adjusts the output level of the corresponding modulator 21B-i individually by adjusting the degree of attenuation to adjust the input level of the optical signal to the optical multiplexer 21D. This is for adjustment, and specifically, each attenuation is adjusted so that the input level of each optical signal to the optical multiplexer 21D becomes uniform. Further, the optical multiplexer (wavelength multiplexing means) 21D multiplexes (n-wavelength multiplexes) the outputs of these variable attenuators 21C-i and outputs (transmits) them to the EDFA 22 as WDM signals.
With such a configuration, in the optical multiplexing unit 21, first, the optical signal (wavelength λ) from the light source 21A- (2k-1) is output by the external modulator 21B- (2k-1). _2k-1 ) Is modulated by the signal Qk, and the external modulator 21B-2k outputs an optical signal (wavelength λ) from the light source 21A-2k. _2k ) Is modulated by the inverted signal Qk.
As a result, for example, as schematically shown in FIGS. 4A and 4B, an optical signal (wavelength λ) having information of the signal Qk is output from the external modulator 21B- (2k-1). _2k-1 ) Is output from the external modulator 21B-2k, and an optical signal (wavelength) having information of a signal Qk obtained by inverting a mark (bit value “1”) and a space (bit value “0”) of the signal Qk is outputted. λ _2k ) Is output.
Each of these optical signals Qk and Qk is adjusted by the corresponding variable attenuator 21C-i so that its optical level becomes uniform at all the wavelengths λ1 to λn, and then is sent to the optical multiplexer 21D. The signal is multiplexed (n-wavelength multiplexed) and output as a WDM signal to the optical fiber transmission line 5-1 through the EDFA 22 and the optical multiplexer 24.
That is, the optical multiplexing unit 21 of the present embodiment outputs the signal Q ₁ And its inverted signal Q ₁ Bars are wavelengths λ1 and λ2, signal Q ₂ And its inverted signal Q ₂ The bars have two different wavelengths λ to transmit a signal Qk (Qk bar) of the same information content, such as wavelengths λ3 and λ4. _2k-1 , Λ _2k It is designed to be sent using. That is, the light source 21A-i, the external modulator 21B-i, and the variable attenuator 21C-i convert the signal Qk and its inverted signal Qk bar into two types of wavelengths λ, as shown in FIG. _2k-1 , Λ _2k Thus, the optical signal generating means 20 for generating the optical signal as described above is configured.
Thereby, these two wavelengths λ _2k-1 , Λ _2k 4A and 4B, the Raman pump light P (wavelength) for “forward pumping” is transmitted through the optical fiber transmission lines 5-1 and 5-2, as schematically shown in FIGS. λ0), but at this time, two wavelengths λ _2k-1 , Λ _2k Assuming that the optical signal propagates while maintaining a substantially synchronous relationship, as shown schematically in FIG. _2k-1 , Λ _2k Is always substantially constant.
As a result, as shown in FIG. 4C, the energy amount of the Raman pumping light P consumed by the signals Qk and Qk bar during Raman amplification becomes uniform, so that the modulation effect on the Raman pumping light P is suppressed. As a result, it is possible to effectively suppress the “crosstalk between channels” that appears remarkably in “forward excitation”.
Here, the phase shift (delay difference) between the signal Qk and its inverted signal Qk due to chromatic dispersion of the optical fiber transmission lines 5-1 and 5-2 serving as Raman amplification media will be estimated. Assuming that a dispersion shift fiber or a non-zero dispersion shift fiber is used for the optical fiber transmission lines 5-1 and 5-2, the optical fiber transmission lines 5-1 and 5-2 have a transmission speed of about 1 ps / km / nm. It will have a variance.
Therefore, for example, assuming that the interval between adjacent channels (wavelengths) is 1 nm and the transmission distance is 100 km (km), the deviation between the signal Qk and its inverted signal Qk due to the dispersion is 100 ps. This delay difference is equivalent to about 3 cm (centimeter) as a transmission path length for one time slot of a signal rate of 10 Gbps.
However, the Raman amplification effect due to “forward pumping” occurs as shown schematically in FIG. 18D, because the portion near the transmitting end is dominant, so that at least at the transmitting end. It is considered that the synchronous relationship is maintained and the crosstalk is effectively suppressed.
However, if there is an optical path difference of one time slot, that is, 3 cm or more before the signal Qk and its inverted signal Qk are combined, the transmitting end will be in a state where no synchronization relationship can be obtained at all. Will be. For this reason, at least in the present optical multiplexing unit 2, it is necessary to wavelength-multiplex the pair of the two signals Qk and Qk bar in the optical multiplexer 21D in a state where they are phase-synchronized with each other.
Therefore, in the present embodiment, at least the optical path length L from the external modulator 21B- (2k-1) to the optical multiplexer 21D, as shown with a circle or a triangle in FIG. _2k-1 And the optical path length L from the external modulator 21B-2k to the optical multiplexer 21D. _2k Are equal to each other, that is, each external modulator 21B- (2k-1) and each external modulator 21B-2k have the same optical path length to the optical multiplexer 21D for each set. The optical path length (arrangement) from 21B-i to the optical multiplexer 21D is designed. Of course, the optical path length between all the external modulators 21B-i and the optical multiplexer 21D may be the same.
Thereby, the set of the signal Qk and the bar Qk can be multiplexed and transmitted by the optical multiplexer 21D in a state where they are phase-synchronized with each other, and the characteristics of the optical multiplexing unit 21 (transmitting station 2) can be improved. Thus, the effect of suppressing "crosstalk between channels" can be maximized.
As described above, according to the transmitting station 2 of the present embodiment, the signal Qk bar in which the waveform of the signal Qk to be transmitted is inverted is generated, and the signal Qk and the signal Qk bar are converted into two adjacent wavelengths λ. _2k-1 , Λ _2k , Which effectively suppresses the "crosstalk between channels" that occurs during Raman amplification due to "forward pumping" without depending on the performance and characteristics of the optical device. As a result, long-distance transmission more than twice that of the related art is possible.
Therefore, if the transmission distance is the same, the number of relay stations required for the system 1 can be greatly reduced as compared with the related art, so that the cost of the system 1 can be reduced. If this is the case, it is possible to construct a system 1 capable of performing long-distance transmission twice or more as compared with the conventional system.
Also, as described above with reference to FIG. 3, in the present embodiment, the inverted signal Qk is obtained by inverting the signal Qk to be transmitted as an electric signal by the inverting gate 21E. _2k Since the inverted optical signal is obtained by modulating the optical signal, the optical signal can be obtained by improving only the electric circuit without making any changes to the basic configuration and optical components of the existing optical transmitter. Since it is completed, it can be easily applied (realized).
Further, since the variable attenuator 21C-i is provided between the external modulator 21-i and the optical multiplexer 21D, the optical level (power) of the signal Qk and its inverted signal Qk bar is reduced. It can be adjusted individually, and it is also possible to control the total power of the signal Qk and its inverted signal Qk bar to an optimum state in which the effect of suppressing the “crosstalk between channels” is maximized.
In the present embodiment, two wavelengths λ are used for a signal Qk (Qk bar) having the same information content. _2k-1 , Λ _2k Inevitably, the use of only one half of the wavelength band compared to the case where one wavelength is assigned to one signal as in the past, and the benefit is reduced. There may be. Hereinafter, this point will be considered.
In general, as a method of increasing the degree of multiplexing in the WDM optical transmission system, there is a method of expanding an amplification band of an optical amplifier and reducing a wavelength interval. As the wavelength interval, for example, a 100 GHz (gigahertz) interval is mainly used in the current device, and in the next-generation new model device, the multiplicity is further increased by setting a half 50 GHz interval or a half 25 GHz interval. It is thought that it is in the direction of raising.
Therefore, as described above, the transmission of the signals Qk and Qk having the same information content requires two wavelengths λ. _2k-1 , Λ _2k In order to achieve long-distance transmission by suppressing the "crosstalk between channels" by adopting the present method using, the wavelength interval may be further reduced in order not to reduce the multiplicity. Now, let's consider how far the wavelength interval can be reduced.
When the wavelength interval is reduced, the factors that limit the transmission characteristics are divided into linear crosstalk and nonlinear crosstalk. Among them, the linear crosstalk may be caused by leakage of the adjacent channel power of the multiplexer / demultiplexer, and this occurs regardless of whether the present method is adopted.
On the other hand, non-linear crosstalk is caused by the self-phase modulation effect (SPM), cross-phase modulation effect (XPM), and four-wave mixing (FWM) in addition to the Raman amplification described above. Here, the wavelength interval between the signal Qk and its inverted signal Qk bar is made as narrow as possible, and it is considered that these two signals Qk and Qk bar appear on almost one wavelength, and the respective powers are summed. Then, power near DC is flowing infinitely. For this reason, the crosstalk that affects other channels from the signal regarded as one wavelength should be close to DC.
Therefore, by adopting this method, it is expected that non-linear crosstalk such as SPM, XPM, and FWM will work in a direction to be suppressed. Further, when the wavelength interval becomes infinitely small, the phase shift between the signals Qk and Qk due to chromatic dispersion also decreases in proportion thereto, so that the crosstalk suppressing effect is expected to be further increased.
Therefore, the signal Qk and its inverted signal Qk of all the wavelengths to be multiplexed are adjacent wavelengths rather than increasing the wavelength multiplexing degree in the conventional method in which different signals are carried on a plurality of wavelengths and transmitted in the conventional method. It is expected that multiplexing of the signals alternately included in the above will easily realize an optical transmission system capable of long-distance transmission with lower noise.
(B) Description of a first modification of the inverted signal generation method
The circuit (optical signal generation means 20) shown in FIG. 3 may have, for example, the configuration shown in FIG. That is, a bias control circuit 213 is provided for each set of the external modulator 21B- (2k-1) and the external modulator 21B-2k, and the external modulator 21B- (2k-1) is used without using the inversion gate 21E. , And 21B-2k are configured to receive the same electric signal Qk.
Then, from the bias control circuit 213, for example, as schematically shown in FIG. 6, the external modulators 21B- (2k-1) and 21B-2k [electrodes (not shown) provided in the optical waveguide of the signal Qk]. By controlling the applied (applied) bias voltage, the light transmittance of the optical waveguide is controlled, and the signal Qk (see the solid line 52) is output from the output port of one of the external modulators 21B- (2k-1). And drives the external modulators 21B- (2k-1) and 21B-2k in a state where the signal Qk is inverted and output (see the broken line 53) from the output port of the other external modulator 21B-2k. I do.
In FIG. 6, a solid line 50 indicates a bias voltage applied to the external modulator 21B- (2k-1), and a broken line 51 indicates a bias voltage applied to the external modulator 21B-2k.
That is, the bias control circuit 213 outputs the signal Qk as an optical signal from one of the external modulators 21B- (2k-1), and outputs the inverted signal Qk as an optical signal from the other external modulator 21B-2k. It functions as a modulation state control circuit that controls the modulation state of each of the external modulators 21B- (2k-1) and 21B-2k so that the bar is output.
Thus, in this case, it is not necessary to invert with an electric signal as described above to obtain the inverted signal Qk bar (there is no need for the inverting gate 21E), so that low cost and downsizing can be achieved. Further, it is possible to avoid occurrence of a delay between the signal Qk and the inverted signal Qk due to a difference in the path of the electric signal (whether the signal passes through the inverting gate 21E or not), thereby preventing the signal Qk from being inverted. The bar can be transmitted in a more synchronized state.
(C) Description of a second modification of the inverted signal generation method
Further, the circuit (optical signal generation means 20) shown in FIG. 3 may have, for example, the configuration shown in FIG. That is, two Mach-Zehnder type optical modulators are arranged in parallel as external modulators 21B- (2k-1) and 21B-2k, and input ports on the same side receive light from the light sources 21A- (2k-1) and 21A-2k. Different wavelengths λ _2k-1 , Λ _2k , And a signal (electric signal) Q to be transmitted to each of the electrodes 211 and 212 is supplied as a modulation signal. FIG. 7 shows a configuration focusing on the external modulators 21B-1 and 21B-2 (wavelengths λ1 and λ2) as a representative example.
Accordingly, if the opposite output ports (in FIG. 7, the output port “2” of the external modulator 21B-1 and the output port “1” of the external modulator 21B-2) are used, the signal Q ₁ And its inverted signal Q ₁ And bar. The operation of the Mach-Zehnder optical modulator itself is known.
7, these two signals Qk and Qk bar (the output port “2” of the external modulator 21B-1 and the output port “1” of the external modulator 21B-2) are optically combined. If the optical modulator 213 is configured to multiplex the signals, the external modulators 21B-1 and 21B-2 and the optical multiplexer 213 are integrated (configured as a Mach-Zehnder type optical modulator / multiplexer) to form a single substrate. It is also possible to be integrated in a.
As described above, by using the Mach-Zehnder type optical modulators for the external modulators 21B- (2k-1) and 21B-2k, the modulator 21-i required for the transmitting station 2 can be made very simple and compact. Therefore, the size of the optical multiplexing unit 21 and thus the transmission station 2 can be significantly reduced. Further, by using the optical multiplexer 213, it is possible to minimize the delay difference between the signal Qk and the inverted signal Qk, and to more effectively exert the effect of suppressing “crosstalk between channels”. be able to.
(D) Description of a third modification of the inverted signal generation method
Next, in order to obtain the inverted signal Qk in the optical signal generating means 20, a method using a semiconductor optical amplifier 21F-k (k = 1 to n), for example, as shown in FIG. 8, may be considered. That is, the wavelength λ from the light source 21A- (2k-1) modulated by the signal to be transmitted is transmitted to the semiconductor optical amplifier 21F-k. _2k-1 And a wavelength λ from another light source 21A-2k. _2k After being multiplexed by the optical multiplexer 215.
That is, the light source 21A- (2k-1) in this case functions as a main signal generation circuit that generates the signal Qk as an optical signal (by a direct modulation method), and the light source 21A-2k functions as a DC signal as an optical signal. Function as a DC signal generation circuit that generates
Then, the gain of the semiconductor optical amplifier 21F-k is controlled by the gain control circuit 214, and the semiconductor optical amplifier 21F-k is operated in a gain saturated state. Then, due to the crosstalk characteristic of the semiconductor optical amplifier 21F-k, the wavelength λ _2k Is modulated. At this time, the inverted signal Qk bar can be obtained by adjusting the gain of the semiconductor optical amplifier 21F-k so that the modulated wave becomes an inverted wave of the signal Qk.
That is, utilizing the fact that the intensity of the DC signal is modulated in accordance with the waveform of the signal Qk by the semiconductor optical amplifier 21F-k (using the semiconductor optical amplifier 21F-k as a modulator), And the inverted signal Qk bar is obtained.
Therefore, also in this case, there is no need to invert the signal with an electric signal, and the semiconductor optical amplifier functioning as a modulator is not adjacent to the adjacent wavelength λ. _2k-1 , Λ _2k Since only one unit is required for each set, cost reduction and miniaturization are possible. Also, no delay occurs between the signal Qk and its inverted signal Qk bar.
Further, in this case, since the input light can be handled as input without converting it into an electric signal, it is possible to configure without using the light source 21A-i. Therefore, for example, an optical signal handled in an optical cross-connect device, an optical ADM (Add-Drop Multiplexer), or the like can be directly handled as an input.
(E) Description of a first modification of the optical multiplexing unit 21
Next, here, a first modified example of the optical multiplexing unit 21 shown in FIG. 2 will be described.
In the optical multiplexing unit 21 described above with reference to FIG. 2, the variable attenuators 21C-i for the number n of wavelengths are provided, but the optical fiber transmission lines 5-1 and 5-2 usually have wavelength-dependent transmission loss characteristics. Therefore, the signal Qk and its inverted signal Qk bar have the adjacent wavelength λ. _2k-1 , Λ _2k , The wavelength λ _2k-1 , Λ _2k It is considered that the difference in transmission loss in the optical fiber transmission lines 5-1 and 5-2 due to the difference is small.
That is, the adjacent wavelength λ _2k-1 , Λ _2k , The transmission loss value to be controlled does not differ greatly, and these adjacent wavelengths λ _2k-1 , Λ _2k It is considered that the deterioration of the characteristics due to the collective control of the optical signals is small. _2k-1 , Λ _2k It is not necessary to control the attenuation (optical transmission power) every time.
Therefore, the signal Qk and its inverted signal Qk are connected to the adjacent wavelength λ. _2k-1 , Λ _2k For example, as shown in FIG. 9, in the optical signal generation means 20, the optical coupler 21G-k is used for each pair of the modulator 21B- (2k-1) and the modulator 21B-2k. The output of the modulator 21B- (2k-1) and the output of the modulator 21B-2k are combined by the optical coupler 21G-k immediately after the output. In the variable attenuator 21C-k, two wavelengths λ _2k-1 , Λ _2k What is necessary is to control collectively the optical signal level of the minute.
However, in this case (or when the configuration described above with reference to FIGS. 7 and 8 is applied), a plurality of (two) wavelengths (channels) λ are provided to the output of one variable attenuator 21C-i. _2k-1 , Λ _2k-1 Therefore, as shown schematically in FIG. 13, for example, two wavelengths λ are included in the pass band per channel. _2k-1 , Λ _2k-1 An optical multiplexer 21D 'having a pass band per channel wider than that of a normal one, which is designed so as to include the optical signals of This makes it possible to further multiplex optical signals including a plurality of wavelengths.
As described above, half of the variable attenuators 21C-k may be provided as compared with the configuration shown in FIG. 2, and as a result, the control of the variable attenuators 21C-k is simplified (that is, the optical transmission power is reduced). Control can be simplified). Therefore, it is possible to greatly reduce the size of the optical multiplexing unit 21, and furthermore, to significantly reduce the size of the transmitting station 2.
Also in this case, as shown by the circles and the triangles in FIG. 9, the optical path length L from the modulator 21B- (2k-1) to the optical coupler 21G-k. _2k-1 'And the optical path length L from the modulator 21B-2k to the optical coupler 21G-k. _2k 'Is designed to be the same.
Thus, also in this case, the pair of the signal Qk and its inverted signal Qk-bar can be multiplexed and transmitted by the optical multiplexer 21D in a state where they are phase-synchronized with each other, thereby suppressing the "crosstalk between channels". Can be maximized. In particular, in this case, the distance (optical path) between the modulators 21B- (2k-1) and 21B-2k and the optical coupler 21G-k, which should have the same optical path length, becomes shorter, so that the signals Qk and Qk It is easy to synchronize the phase with the bar, and the design is easy.
The above configuration can be applied to, for example, a transmitting station (optical multiplexing unit 21 ') in a normal WDM optical transmission system in which different signals are loaded on wavelengths to be multiplexed and transmitted, as shown in FIG. Good.
That is, even in a normal WDM optical transmission system, the adjacent wavelength λ _2k-1 , Λ _2k Since the difference in transmission loss is negligible, the optical signal from the light source 21A-i is converted into a different signal (transmission data) Q by the corresponding modulator 21B-i. ₁ Also, in the case of a configuration in which modulation is performed by Qn, an optical coupler 21G-k is provided for each set of the modulators 21B- (2k-1) and 21B-2k, and the modulators 21B- (2k-1) are provided. And the output of the modulator 21B-2k are coupled by the optical coupler 21G-k.
As a result, in the optical multiplexing unit 21 'applied to the transmitting station of the ordinary WDM optical transmission system, the optical transmission power of a plurality of channels (wavelengths) is not divided by channel but by half of the variable attenuators 21C-k. Adjacent wavelength λ _2k-1 , Λ _2k It can be controlled collectively every time.
Therefore, also in this case, the number of variable attenuators 21C-k can be saved, and the circuit for controlling each variable attenuator 21C-k can be reduced. Cost reduction and downsizing are possible, and its stability is also improved. Also in this case, since the optical path from the modulator 21B-i to the optical multiplexer 21D is short, the adjacent wavelength λ _2k-1 , Λ _2k It becomes easy to multiplex while maintaining the phase synchronization between them.
(F) Description of a second modification of the optical multiplexing unit 21
The optical multiplexing unit 21 (optical signal generation unit 20) described above with reference to FIG. 2 (or FIG. 9) may have a configuration as shown in FIG. 11, for example.
That is, a serial / parallel (S / P) converter 216 that serially / parallel converts a signal Qk to be transmitted and reduces the signal speed (for example, 10 Gbps) to （(5 Gbps), and this S / P converter A selector 217 for selecting one of the output (half) of the signal 216 and the signal Qk and outputting the selected signal as a modulation signal of the modulator 21B- (2k-1); and the output (the other half) of the S / P converter 216. A selector 218 that selects one of the outputs of the inverting gate 21E and outputs the selected signal as a modulation signal of the modulator 21B-2k is connected to the adjacent wavelength λ. _2k-1 , Λ _2k May be provided for each set.
That is, the S / P conversion section 216 functions as a transmission rate conversion section that converts the transmission rate of the signal Qk, and the selectors 217 and 218 operate as a set of the signal Qk and the output of the inverting gate 21E or S / P conversion. The selector 216 functions as a selector for selecting one of the outputs and inputting it to each of the modulators 21B- (2k-1) and 21B-2k. Each of the selectors 217 and 218 sets a signal to be selected by, for example, external setting.
In the optical multiplexing unit 21 configured as described above, by switching the outputs of the selectors 217 and 218 in accordance with the required transmission band and the conditions of the optical fiber transmission lines 5-1 and 5-2, the signal Qk and its inversion are switched. The signal Qk and the two wavelengths λ are kept at 10 Gbps. _2k-1 , Λ _2k And the two types of wavelengths λ by reducing only the signal Qk to 5 Gbps. _2k-1 , Λ _2k Can be switched to the "speed conversion mode" to be transmitted.
Accordingly, when the dispersion of the optical fiber transmission lines 5-1 and 5-2 is large and it is difficult to suppress waveform deterioration, or when it is difficult to transmit a high-speed signal due to large nonlinearity, the latter "speed" When Raman amplification is used to cope with long transmission distances (relay distances) using the "conversion mode", various characteristics of the optical transmission line, such as using the former "crosstalk suppression mode" It is possible to provide a high value-added device (transmitting station 2) that can respond to customer requests such as upgrading from the introduced configuration.
When the optical multiplexing unit 21 of the transmitting station 2 has the above configuration, the optical demultiplexing unit 44 of the receiving station 4 also selects one of the above “crosstalk suppression mode” and “speed conversion mode”. A configuration that can be used. The details will be described later with reference to FIG.
(G) Description of the optical demultiplexing unit 44 of the receiving station 4
Next, FIG. 14 is a block diagram showing the configuration of the optical demultiplexing unit 44 in the receiving station 4. The optical demultiplexing unit 44 shown in FIG. 14 includes an optical demultiplexer 44A and a band-pass filter (BPF: Band). Pass Filter) 44B-1 to 44B-n, optical receivers 44C-1 to 44C-n, characteristic monitoring unit 44D-k (k = 1 to n / 2), inverting gate 44E-k, and selector (SEL) 44F- k.
Here, the optical demultiplexer 44A is for demultiplexing the optical signal (WDM signal) from the optical fiber transmission line 5-2 pre-amplified by the EDFA 43 into optical signals of the respective wavelengths λ1 to λn. Then, for example, an arrayed waveguide grating filter is applied. Further, the BPFs 44B-i are for passing only the optical signal of the wavelength λi component to remove unnecessary components such as noise components, and the optical receivers 44C-i respectively correspond to the respective components. This is for performing a receiving process (such as photoelectric conversion) on the optical signal from the BPF 44B-i.
The characteristic (quality) monitoring units 44D-k respectively transmit the wavelengths λ received by the optical receivers 44C- (2k-1). _2k-1 Electrical signal (signal Q) and the wavelength λ received by the optical receiver 44C-2k _2k Of each electrical signal (inverted signal Qk bar of signal Q) (waveform, signal error rate (bit error rate), etc.) by monitoring _2k-1 , Λ _2k The inverting gates 44E-k are for inverting the signal Qk received by the optical receivers 44C-2k to obtain the original signal Qk.
Each of the selectors 44F-k outputs the wavelength λ from the optical receiver 44C- (2k-1). _2k-1 And the wavelength λ from the inverting gate 44E-k _2k In the present embodiment, a signal having better signal quality is selected as a reception signal by a selection control signal according to the monitoring result of the characteristic monitoring unit 44D-k. It is supposed to.
In the optical demultiplexing unit 44 of the present embodiment configured as described above, the WDM signal from the EDFA 43 is demultiplexed into the optical signals of the respective wavelengths λ1 to λn by the demultiplexing 44A, and then the respective signals are separated by the BPF 44B-i. Unnecessary components such as noise components are removed, received by the optical receiver 44C-i, and converted into electric signals.
At this time, in the characteristic monitoring unit 44D-k, the wavelength λ received by the optical receiver 44C- (2k-1) is used. _2k-1 And the wavelength λ received by the optical receiver 44C-2k. _2k Of the inverted signal Qk is monitored by calculating a bit error rate and the like, and the selector 44F-k is controlled so that the better signal quality is selected.
As a result, a high quality wavelength λ _2k-1 Or λ _2k Is selected as the signal of the working line (channel).
As described above, in the receiving station 4 (the optical demultiplexing unit 44) of the present embodiment, the transmitting station 2 has a plurality of wavelengths λ. _2k-1 , Λ _2k Utilizing the fact that a signal having the same information content is transmitted using, the signal that is received with better signal quality is selected as the signal of the working channel, so that better transmission characteristics can be guaranteed.
Further, even in a case where an abnormality such as a failure occurs in some of the light sources 21A-i for the wavelength λi in the transmitting station 2, and the receiving power of the some wavelength λi is reduced in the receiving station 4, , Normal reception can be performed by other wavelengths forming a pair, so that security and reliability close to duplexing of lines (channels) can be obtained.
(H) Description of First Modification of Optical Demultiplexing Unit 44
Next, FIG. 15 is a block diagram showing a first modification of the above-described optical demultiplexing unit 44. The optical demultiplexing unit 44 shown in FIG. 15 has the same optical demultiplexer 44A as that described above with reference to FIG. , BPFs 44B-1 to 44B-n and optical receivers 44C-1 to 44C-n, and a differential amplifier 44G-k (k = 1 to n / 2).
Here, these differential amplifiers 44G-k are connected to the electric signal (Qk) from the optical receiver 44C- (2k-1) and the electric signal (inverted signal Qk bar) from the optical receiver 44C-2k, respectively. , And by detecting the difference between them, the DC component of the transmission line noise can be canceled (cancelled), similarly to the principle of the differential amplifier for removing the common-mode noise in the transmission line of the electric signal.
With this configuration, the optical demultiplexer 44 cancels in-phase noise components such as ASE (Amplified Spontaneous Emission) generated in the optical fiber transmission lines 5-1 and 5-2 by the differential amplifier 44G-k. Therefore, a better signal-to-noise ratio can be realized, and a longer relay distance can be handled.
(I) Description of Second Modified Example of Optical Demultiplexing Unit 44
FIG. 16 is a block diagram showing a second modification of the above-described optical demultiplexing unit 44. The optical demultiplexing unit 44 shown in FIG. 16 includes the optical demultiplexing unit 21 of the transmitting station 2 as described above with reference to FIG. This corresponds to the receiving side when a function of switching between the “crosstalk suppression mode” and the “speed conversion mode” is added. The optical demultiplexer 44A is the same as that described above, and the BPF 44B-1 for each of the wavelengths λ1 to λn. 44B-n, optical receivers 44C-1 to 44C-n for each of the wavelengths λ1 to λn, and an inverted wave receiving circuit 441, a parallel / serial (P / S) converter 442 and a selector 443 (See FIG. 11).
Here, the inverted wave receiving circuit 441 is, for example, a circuit including the characteristic monitoring unit 44D-k, the inverting gate 44E-k, and the selector 44F-k described above with reference to FIG. 14, or the differential amplifier described above with reference to FIG. The circuit corresponding to 44G-k receives the output of the optical receiver 44C- (2k-1) and the output of the optical receiver 44C- (2k-1) as inputs.
Therefore, if the transmitting station 2 is set to the “crosstalk suppression mode”, the inverted wave receiving circuit 441 has the wavelength λ _2k-1 Qk and wavelength λ sent by _2k In this case, the inverted signal Qk bar sent in the step (1) is input, and if the “speed conversion mode” is set, the speed is converted by the S / P converter 216 on the transmitting station 2 side (reduced by half). 2) wavelength λ _2k-1 , Λ _2k Will be input.
Further, P / S conversion section 442 receives as inputs the output of optical receiver 44C- (2k-1) and the output of optical receiver 44C- (2k-1), and performs P / S conversion on the input signal. (Speed conversion), for performing speed conversion (double when the transmission side 2 reduces the frequency by half) according to the speed conversion by the S / P conversion unit 216 of the transmitting station 2. .
The selector 443 sets a mode according to the mode setting on the transmitting station 2 side, and selects one of the output of the inverted wave receiving circuit 441 and the output of the P / S converter 442 in accordance with the setting. For example, in the case of “crosstalk suppression mode”, the output of the inverted wave receiving circuit 441 is selected, and in the case of “velocity conversion mode”, the output of the P / S converter 442 is selected. ing.
In the optical demultiplexer 44 configured as described above, in the “crosstalk suppression mode”, the output of the inverted wave receiving circuit 441 becomes effective, and the wavelength λ _2k-1 Qk and wavelength λ sent by _2k And the difference detection result by the differential amplifier 44G-k is output from the inverted signal Qk bar sent in the above. In the "speed conversion mode", the output of the P / S converter 442 is valid. The speed is reduced on the transmitting station 2 side (for example, 5 Gbps) and the two wavelengths λ _2k-1 , Λ _2k Is output at an increased speed (for example, 10 Gbps).
In this way, by operating according to the mode setting on the transmitting station 2 side, similarly to the transmitting side, added value that can respond to customer requirements such as upgrading from various optical transmission line characteristics and the initially introduced configuration. (The receiving station 2) having a high power consumption.
(J) Other
By the way, when multi-stage optical amplification relay is performed as schematically shown in FIG. 18A, the inverted signal Qk having a different wavelength from the above signal Qk due to the chromatic dispersion characteristic of the optical fiber transmission line 5. As schematically shown in FIG. 18B, delays are accumulated between the bar and the bar as the transmission (relay) distance increases. On the other hand, as described above with reference to FIG. 18D, the Raman amplification effect due to “forward pumping” is greatest immediately after the output of the transmission station 2 and the relay station 3 (transmission end).
Therefore, in order to effectively suppress “inter-channel crosstalk” in the relay station 3 as well, it is desirable that the relay station 3 also compensates for the delay between the signal Qk and its inverted signal Qk bar. Therefore, for example, as shown in FIG. 17, at least the relay station 2 has a dispersion compensating fiber (DCF) as a dispersion compensator having a dispersion value that compensates for the chromatic dispersion characteristic of the optical fiber transmission line 5 in front of it. : Dispersion Compensating Fiber) 304 is provided. Note that the DCF 304 has a limit on the input light power (a noise component increases if the input light power is too high), and thus is usually provided before the EDF 301.
Thus, as schematically shown in FIG. 18C, the delay between the signal Qk and its inverted signal Qk can be reduced immediately after the output of the relay station 3. As a result, even in the system 1 that performs multi-stage optical amplification relay, it is possible to effectively bring out the effect of suppressing “crosstalk between channels” over the entire transmission distance only by providing the DCF 304 in the relay station 3.
However, since “Raman amplification” uses the optical fiber transmission line 5 itself, which is a very long distance of several km to several tens km, as an amplifying medium, a method of generating crosstalk due to the dispersion characteristics and loss characteristics of the optical fiber transmission line 5. The transmission characteristics may not always be the best when the signal Qk and its inverted signal Qk are transmitted in a completely synchronized state (a state where the delay difference is zero) as described above.
Therefore, for example, as schematically shown in FIGS. 19 and 7, an electrode 221 is provided on a path (a dielectric optical waveguide or the like) through which an inverted optical signal Qk bar (or signal Qk) passes, and the electrode 221 has a refractive index. A voltage may be applied from the control circuit (timing control circuit) 222 to control the refractive index of light, so that the optical path length of the inverted signal Qk (or the signal Qk) may be adjusted.
Thereby, the delay difference Δτ between the signal Qk and its inverted signal Qk bar, that is, the output timing of the signal Qk and its inverted signal Qk bar can be appropriately adjusted. Therefore, even after the system operation is started, the transmission characteristics can be optimized by adjusting the delay difference Δτ including the one caused by the temperature change and the aging, and the crosstalk suppression effect can always be obtained. You can make the most of it.
In the above-described embodiment, two adjacent wavelengths λ _2k-1 , Λ _2k Has been described, the crosstalk is suppressed by transmitting the signal Qk (Qk bar) having the same information content. However, even when three or more wavelengths are used, the crosstalk suppression effect can be obtained in the same manner as in the above-described embodiment. Obtainable.
For example, taking the case of three wavelengths as an example, as shown in FIGS. 20A and 20B, the signal Qk has a wavelength λ. _2k , The inverted signal Qk has a wavelength λ _2k-1 And wavelength λ _{2k + 1} Are transmitted at half the level (power) of the signal Qk. Thus, also in this case, three adjacent wavelengths λ _2k-1 , Λ _2k , Λ _{2k + 1} If each optical signal is wavelength-division multiplexed and transmitted while maintaining a synchronous relationship, the total optical power becomes constant, so that the modulation effect on the Raman pump light is suppressed, and crosstalk can be suppressed.
Furthermore, in the above-described embodiments, the suppression of crosstalk during “Raman amplification” has been described consistently. However, even when a semiconductor optical amplifier is used, the same operation and effect as in the above-described embodiments can be obtained.
That is, if the signal Qk and its inverted signal Qk are input to the semiconductor optical amplifier in synchronization with each other, the total power of each signal Qk and Qk becomes constant, so that the carrier density in the active region in the semiconductor optical amplifier is reduced. The fluctuation is suppressed, and as a result, the signal fluctuation due to the fluctuation in the gain and the “pattern effect” is also suppressed, and the crosstalk can be effectively suppressed.
The inverted signal Qk does not necessarily have to have a completely inverted waveform of the signal Qk. That is, for example, even if the inverted signal Qk and the signal Qk have slightly different optical powers and waveform deviations, their total powers are substantially constant as a whole, so that a sufficient crosstalk suppression effect can be obtained. It is thought that it is possible.
Further, in the above-described embodiment, the external modulation method of externally modulating the optical signal from the light source 21A-i with the signal Qk or Qk bar is adopted, but the signal Qk or Qk bar is directly input to the light source 21A-i. Alternatively, a direct modulation method for performing modulation may be employed.
In the above-described embodiment, the case where the present invention is applied to the hybrid system in which the EDFA 22 (33, 43) and the Raman amplifier (or the semiconductor optical amplifier) are combined has been described. Even when applied to a WDM optical transmission system using a single unit, the same operation and effect as described above can be obtained.
Further, in the above-described embodiment, the signal Qk and its inverted signal Qk are connected to the adjacent wavelength λ. _2k-1 , Λ _2k Is transmitted using the adjacent wavelength λ. _2k-1 , Λ _2k Sometimes it is not necessary to use. For example, when a semiconductor optical amplifier is used instead of a Raman amplifier, the active region where an optical signal is amplified is about several hundred μm to 1 mm. It is considered that there is almost no influence of delay due to dispersion. Therefore, when a semiconductor optical amplifier is used, it is not necessary to use adjacent wavelengths, and it is considered that any wavelength within the gain band can be used.
Further, in the above-described embodiment, the case where the “bidirectional pumping” Raman amplification is applied to the WDM optical transmission system 1 has been described. Of course, the same applies to the case where only the “forward pumping” is applied. The operation and effect of the invention can be obtained. In the above-described embodiment, all the signals Qk to be transmitted are transmitted as a set with the inverted signal Qk, but only some of the signals Qk are transmitted as a set with the inverted signal Qk. You may make it transmit.
For example, if a predetermined distance can be transmitted with sufficient signal quality only by transmitting a part of the signal Qk in combination with its inverted signal Qk bar, the remaining signal Qk can be transmitted without using the inverted signal Qk bar. Normal transmission may be performed. For wavelengths that have optical power that easily causes crosstalk to affect other wavelengths (channels) due to the wavelength-dependent loss characteristics of the optical transmission line and the optical amplifier, the signals are transmitted in pairs with the inverted signal Qk. The other wavelengths may be transmitted without using the inverted signal Qk.
In this way, even if the optical power varies depending on the wavelength due to the wavelength-dependent loss characteristic of the optical transmission line or the optical amplifier, the influence of the crosstalk can be suppressed.
The present invention is not limited to the above-described embodiment, and may be variously modified and implemented without departing from the spirit of the present invention.
Industrial applicability
As described above, according to the present invention, in a wavelength-division multiplexing optical transmission system, crosstalk between channels that appears remarkably when Raman amplification of “forward pumping” is used is independent of the performance and characteristics of an optical device. Since the suppression can be performed effectively, the wavelength multiplexed optical signal can be transmitted over a long distance with lower noise than before, and its usefulness is considered to be extremely high.
[Brief description of the drawings]
FIG. 1 is a block diagram showing a configuration of a wavelength division multiplexing (WDM) optical repeater transmission system according to an embodiment of the present invention.
FIG. 2 is a block diagram showing a configuration of the optical multiplexing unit in the transmitting station shown in FIG.
FIG. 3 is a block diagram showing a configuration focusing on the light source and the modulator shown in FIG.
FIG. 4A is a schematic diagram illustrating a wavelength (channel) arrangement example of the Raman pump light, the signal to be transmitted, and its inverted signal according to the present embodiment.
FIG. 4B is a schematic diagram illustrating an example of the waveform of the Raman pump light before Raman amplification, the signal to be transmitted, and its inverted signal according to the present embodiment.
FIG. 4C is a schematic diagram illustrating a waveform example of the Raman pumping light after Raman amplification, a signal to be transmitted, and its inverted signal according to the present embodiment.
FIG. 5 is a block diagram for explaining a first modification of the inverted signal generation method according to the present embodiment.
FIG. 6 is a schematic diagram for explaining a bias control method for the modulator shown in FIG.
FIG. 7 is a block diagram for explaining a second modification of the inversion signal generation method according to the present embodiment.
FIG. 8 is a block diagram for explaining a third modification of the inverted signal generation method according to the present embodiment.
FIG. 9 is a block diagram showing a first modification of the optical multiplexing unit shown in FIGS.
FIG. 10 is a block diagram for explaining that the optical multiplexing unit shown in FIG. 9 can be applied to a normal WDM optical transmission system.
FIG. 11 is a block diagram showing a second modification of the optical multiplexing unit shown in FIGS.
FIG. 12 is a block diagram showing the configuration of the EDFA shown in FIG.
FIG. 13 is a schematic diagram showing an example of the pass band characteristics of the optical multiplexer shown in FIG. 9 (or FIG. 10).
FIG. 14 is a block diagram showing the configuration of the optical demultiplexing unit in the receiving station shown in FIG.
FIG. 15 is a block diagram showing a first modification of the optical branching unit shown in FIG.
FIG. 16 is a block diagram showing a second modification of the optical branching unit shown in FIG.
FIG. 17 is a block diagram showing a modification of the EDFA shown in FIG.
FIG. 18A is a block diagram showing a WDM optical transmission system when performing multistage optical amplification relay.
FIG. 18B is a schematic diagram showing a delay difference according to a transmission distance between a transmission signal and its inverted signal in the case where no DCF is provided in the system shown in FIG. 18A.
FIG. 18C is a schematic diagram showing a delay difference according to a transmission distance between a transmission signal and its inverted signal when a DCF is provided in a relay station in the system shown in FIG. 18A.
FIG. 18D is a schematic diagram illustrating a Raman gain according to a transmission distance by Raman amplification of “forward excitation” in the system illustrated in FIG. 18A.
FIG. 19 is a schematic diagram for explaining control of a delay difference between a signal to be transmitted and its inverted signal according to the present embodiment.
FIG. 20A and FIG. 20B are schematic diagrams for explaining a case where three wavelengths are used for transmitting a signal to be transmitted and its inverted signal according to the present embodiment.
FIG. 21 is a block diagram showing an example of a conventional WDM optical transmission system using both an EDFA and a Raman amplifier.
FIG. 22 is a schematic diagram for explaining a relay gain and a spontaneous emission optical noise in a WDM optical transmission system using a conventional EDFA and a Raman amplifier.
FIGS. 23A and 23B are schematic diagrams for explaining a modulation effect on Raman pumping light at the time of Raman amplification.
FIG. 24A is a schematic diagram illustrating a wavelength (channel) arrangement example of Raman pump light and two signals to be transmitted.
FIG. 24B is a schematic diagram illustrating an example of waveforms before Raman amplification of the Raman pump light and the two signal lights to be transmitted illustrated in FIG.
FIG. 24C is a schematic diagram illustrating an example of waveforms after Raman amplification of the Raman pump light and the two signal lights to be transmitted illustrated in FIG.
FIG. 25A is a block diagram showing a "forward pumping" Raman amplifier configuration.
FIG. 25B is a block diagram showing the configuration of the “backward pumping” Raman amplifier.
FIG. 25C is a block diagram showing a “bidirectional pump” Raman amplifier configuration.
26A to 26C are schematic diagrams for explaining the “pattern effect” of the semiconductor optical amplifier.
FIGS. 27A to 27E are schematic diagrams for explaining “inter-channel crosstalk” due to the “pattern effect” of the semiconductor optical amplifier.

Claims (21)

An optical signal generating means (20) for generating a main signal to be transmitted and its inverted signal as optical signals of a plurality of wavelengths;
An optical transmitter, comprising: wavelength multiplexing means (21D) for multiplexing and transmitting the optical signals of the plurality of wavelengths generated by the optical signal generating means. The optical signal generation means (20)
2. The optical transmitter according to claim 1, wherein the main signal and the inverted signal are output in a synchronized state. The optical signal generation means (20)
An inverting circuit (21E) for inverting the main signal as an electric signal;
A first light source [21A- (2k-1); which generates an optical signal of a certain wavelength; k = 1 to n / 2, and n is an even number of 2 or more]
A second light source (21A-2k) for generating an optical signal having a wavelength different from the wavelength of the optical signal generated by the first light source;
A first modulator [21B- (2k-1)] for modulating an optical signal from the first light source [21A- (2k-1)] with the main signal;
A second modulator (21B-2k) for modulating an optical signal from the second light source (21A-2k) with an output of the inverting circuit (21E). The optical transmitter according to claim 1. The optical signal generation means (20)
A first light source [21A- (2k-1)] for generating an optical signal of a certain wavelength;
A second light source (21A-2k) for generating an optical signal having a wavelength different from the wavelength of the optical signal generated by the first light source;
First and second modulators [21B- (2k-1)] that modulate the optical signals from the light sources [21A- (2k-1) and 21A-2k] with the same main signal as an electric signal. , 21B-2k],
The above-mentioned arrangement is such that the main signal as an optical signal is output from one modulator [21B- (2k-1)] and the inverted signal as an optical signal is output from the other modulator (21B-2k). 2. A modulator according to claim 1, further comprising a modulation state control circuit (213) for controlling a modulation state of each modulator [21B- (2k-1), 21B-2k]. Optical transmitter. The optical signal generation means (20)
An optical multiplexer (215) for multiplexing the main signal as an optical signal and the DC signal as an optical signal;
2. The optical transmitter according to claim 1, further comprising a semiconductor optical amplifier (21F-k) to which an output of said optical multiplexer (215) is input. The optical path length from the first modulator [21B- (2k-1)] to the wavelength multiplexing means (21D) and the optical path length from the second modulator (21B-2k) to the wavelength multiplexing means (21D). The optical transmitter according to claim 3 or 4, wherein the length is the same. The optical signal generation means (20)
A variable attenuator (21C-i; i = 1 to n) for adjusting the output level of each of the modulators [21B- (2k-1), 21B-2k] is provided. The optical transmitter according to claim 6, wherein: The optical signal generation means (20)
An optical coupler (21G-k) for coupling the outputs of the first and second modulators [21B- (2k-1), 21B-2k];
The optical path length from the first modulator [21B- (2k-1)] to the optical coupler (21G-k) and the optical path length from the second modulator (21B-2k) to the optical coupler (21G-k) The optical transmitter according to claim 3 or 4, wherein the optical path lengths up to (3) are the same. 9. The optical signal generating means according to claim 8, wherein said optical signal generating means includes a variable attenuator for adjusting an output level of said optical coupler. Optical transmitter. The optical signal generation means (20)
A transmission rate converter (216) for converting the transmission rate of the main signal;
The modulator [21B- (2k-1), 21B- 2k]. The optical transmitter according to claim 3, further comprising a selection unit (217, 218) for inputting the input signal to the optical transmitter. Each of the modulators [21B- (2k-1), 21B-2k] is configured as a Mach-Zehnder type optical modulator / multiplexer for multiplexing outputs from different output ports of two Mach-Zehnder type optical modulators. 5. The optical transmitter according to claim 4, wherein: The wavelength multiplexing means (21D) is configured using an optical multiplexer (21D ') having the plurality of types of wavelengths as passbands per channel, wherein: Item 9. The optical transmitter according to Item 8. The optical signal generation means (20)
The optical transmitter according to claim 1, further comprising a timing control circuit (222) for controlling the output timing of the main signal and the inverted signal. The optical transmitter according to any one of claims 1 to 13, wherein the plurality of types of wavelengths are adjacent wavelengths. A plurality of light sources (21A-i) each generating an optical signal having a different wavelength,
A modulator (21B-i) provided for each of the light sources (21A-i) and modulating an optical signal from the light source (21A-i) with a main signal to be transmitted;
An optical coupler (21G-k) that couples the output of the modulator (21B-i) at least for every two adjacent pairs of wavelengths;
A variable attenuator (21C-k) for adjusting an output level of the optical coupler (21G-k);
An optical transmitter comprising an optical multiplexer (21D ') for multiplexing the output of the variable attenuator (21C-k). An optical repeater (3) for relaying an output of an optical transmitter (2) for transmitting a main signal to be transmitted and an inverted signal thereof as wavelength-multiplexed optical signals of a plurality of wavelengths,
An optical repeater comprising a dispersion compensator (304) for compensating chromatic dispersion between the main signal and the inverted signal. An optical receiver (4) for receiving an output of an optical transmitter (2) for transmitting a main signal to be transmitted and an inverted signal thereof as wavelength-multiplexed optical signals of a plurality of wavelengths,
A quality monitoring unit (44D-k) for respectively monitoring the quality of the main signal and the inverted signal;
A selection unit (44F-k) for selecting one of the main signal and the inverted signal as a reception signal according to the quality monitoring result of the quality monitoring unit (44D-k). Characteristic, optical receiver. An optical receiver (4) for receiving an output of an optical transmitter (2) for transmitting a main signal to be transmitted and an inverted signal thereof as wavelength-multiplexed optical signals of a plurality of wavelengths,
An optical splitter (44A) for splitting the wavelength multiplexed optical signal into the main signal and the inverted signal;
An optical receiver comprising a differential amplifier (44G-k) to which the main signal and the inverted signal from the optical splitter (44A) are input. An optical transmission method, wherein a main signal to be transmitted and its inverted signal are transmitted as optical signals of a plurality of wavelengths. 20. The optical transmission method according to claim 19, wherein the main signal and the inverted signal are transmitted in a synchronized state. 21. The optical transmission method according to claim 19, wherein the plurality of types of wavelengths are adjacent wavelengths.

Images